Failure of Prestressed Concrete Cylinder Pipe

Failure of Prestressed
Concrete Cylinder Pipe
Subject Area:
Infrastructure Reliability
2008 AwwaRF. ALL RIGHTS RESERVED
About the Awwa Research Foundation

The Awwa Research Foundation (AwwaRF) is a member-supported, international, nonprofit organization
that sponsors research to enable water utilities, public health agencies, and other professionals to provide
safe and affordable drinking water to consumers.
The Foundations mission is to advance the science of water to improve the quality of life. To achieve
this mission, the Foundation sponsors studies on all aspects of drinking water, including supply and
resources, treatment, monitoring and analysis, distribution, management, and health effects. Funding
for research is provided primarily by subscription payments from approximately 1,000 utilities, consulting
firms, and manufacturers in North America and abroad. Additional funding comes from collaborative
partnerships with other national and international organizations, allowing for resources to be leveraged,
expertise to be shared, and broad-based knowledge to be developed and disseminated. Government
funding serves as a third source of research dollars.
From its headquarters in Denver, Colorado, the Foundations staff directs and supports the efforts of
more than 800 volunteers who serve on the board of trustees and various committees. These volunteers
represent many facets of the water industry, and contribute their expertise to select and monitor research
studies that benefit the entire drinking water community.
The results of research are disseminated through a number of channels, including reports, the Web site,
conferences, and periodicals.
For subscribers, the Foundation serves as a cooperative program in which water suppliers unite to pool
their resources. By applying Foundation research findings, these water suppliers can save substantial
costs and stay on the leading edge of drinking water science and technology. Since its inception, AwwaRF
has supplied the water community with more than $300 million in applied research.
More information about the Foundation and how to become a subscriber is available on the Web
at www.awwarf.org.
Prepared by:
Andrew E. Romer and Dan Ellison
Boyle Engineering Corporation
1501 Quail Street, Newport Beach, CA 92660-2726
and
Graham E. C. Bell and Brien Clark
Schiff Associates, Inc.
431 West Baseline Road, Claremont, CA 91711
Jointly sponsored by:

Awwa Research Foundation
6666 West Quincy Avenue, Denver, CO 80235-3098
and
U.S Environmental Protection Agency
Washington, DC
Published by:
Distributed by:
DISCLAIMER
This study was jointly funded by the Awwa Research Foundation (AwwaRF) and the U.S. Environmental Protection
Agency (USEPA) under Cooperative Agreement No. CR826432-01. AwwaRF and USEPA assume no responsibility
for the content of the research study reported in this publication or for the opinions or statements of fact expressed
in the report. The mention of trade names for commercial products does not represent or imply the approval
or endorsement of either AwwaRF or USEPA. This report is presented solely for informational purposes.
Copyright 2008
by Awwa Research Foundation
ALL RIGHTS RESERVED.
No part of this publication may be copied, reproduced
or otherwise utilized without permission.
ISBN 978-1-60573-013-4
Printed in the U.S.A.
CONTENTS
TABLES ....................................................................................................................................... vii
FIGURES....................................................................................................................................... ix
FOREWORD ............................................................................................................................... xiii
ACKNOWLEDGMENTS .............................................................................................................xv
EXECUTIVE SUMMARY ........................................................................................................ xvii
CHAPTER 1 INTRODUCTION .....................................................................................................1
Literature Search..................................................................................................................1
Research to Identify PCCP with High Likelihood of Failure........................................1
Alleviating PCCP Failures.............................................................................................2
Cost of Failures ..............................................................................................................2
Survey of Utilities to Supplement Existing Database....................................................3
Predicting Failures .........................................................................................................3
CHAPTER 2 HISTORY OF PCCP ...............................................................................................53
Production Data .................................................................................................................53
Changes in Pipe Standards.................................................................................................55
Size Range ...................................................................................................................60
Design Basis ................................................................................................................60
Concrete Quality ..........................................................................................................63
Steel for Cylinders .......................................................................................................64
Steel Cylinder Thickness .............................................................................................64
Steel Cylinder Weld Tests ...........................................................................................65
Wire Reinforcement.....................................................................................................65
Wire Wrapping Stress ..................................................................................................65
Wire Reinforcement Spacing.......................................................................................65
Concrete Core Manufacturing Tolerances...................................................................66
Joint Ring Tolerances ..................................................................................................66
Strength of Concrete ....................................................................................................67
Core Prestressing Limits ..............................................................................................67
Concrete Coating .........................................................................................................68
Summary ......................................................................................................................68
Changes in Material Standards ..........................................................................................69
Changing References to ASTM Wire Standards .........................................................73
Changes Within the ASTM Wire Standards................................................................74
Commercial Changes Beyond the ASTM Wire Standards..........................................75
v
CHAPTER 3 PERFORMANCE....................................................................................................83
Causes and Modes of Failure.............................................................................................83
Definition of PCCP Failure..........................................................................................83
Common Causes of PCCP Failure ...............................................................................83
How PCCP Fails ..........................................................................................................86
Design Life.........................................................................................................................87
Definition of Expected Service Life for PCCP............................................................87
Data Partitions....................................................................................................................87
Classes of Failure.........................................................................................................87
Failure Database ................................................................................................................89
Histograms and Simple Statistical Analysis ......................................................................89
PCCP Failures by Age .................................................................................................89
PCCP Failures by Installation Date .............................................................................92
PCCP Failure Rates by Manufacturing Era .................................................................93
PCCP Failures by Wire Class ......................................................................................93
PCCP Failure Rates by Pipe Type by Installation Year ..............................................94
CHAPTER 4 ASSESSMENT MODEL.......................................................................................115
Assessment Methods Types and Results ......................................................................115
Workshop With Utilities ..................................................................................................115
Assessment (Evaluation) Matrices...................................................................................116
PCCP Data Weighting Factors...................................................................................116
Short-Form Assessment Matrix .................................................................................118
Long-Form Assessment Matrix .................................................................................119
Validation of Assessment Matrices .................................................................................124
CHAPTER 5 PREDICTING SERVICE LIFE.............................................................................125
Ogives and Failure Forecasting .......................................................................................125
Weibull Analysis and Failure Prediction .........................................................................128
Probability of PCCP Failure by Era.................................................................................135
Category 3 Failure Prediction Interference................................................................137
CHAPTER 6 CONCLUSIONS ...................................................................................................173
Expected Service Life for PCCP......................................................................................175
CHAPTER 7 FUTURE STUDY RECOMMENDATIONS........................................................177
APPENDIX A: QUALITATIVE ASSESSMENT MATRIX......................................................179
Appendix A1: Short Form ...............................................................................................180
Appendix A2: Long Form................................................................................................181
GLOSSARY ................................................................................................................................185
REFERENCES AND BIBLIOGRAPHY ....................................................................................195
ABBREVIATIONS .....................................................................................................................201
vi
TABLES
1.1
Reported failures of prestressed concrete cylinder pipe ......................................................5
2.1
Differences between lined and embedded type PCCP.......................................................53
2.2
PCCP significant changes with time..................................................................................53
2.3
Production of prestressed concrete cylinder pipe ..............................................................56
2.4
AWWA C301 prestressed concrete cylinder pipe timeline ...............................................61
2.5
Minimum wire clearance for minimum wire size..............................................................66
2.6
Joint dimensions ................................................................................................................67
2.7
Timeline of wire standards for PCCP ................................................................................69
3.1
Sample populations............................................................................................................89
5.1
Summary statistics ...........................................................................................................129
5.2
Weibull summary statistics ..............................................................................................136
vii
FIGURES
1.1
Location of failures LC-PCCP........................................................................................51
1.2
Location of failures EC-PCCP........................................................................................52
2.1
Timeline ................................................................................................................. see insert
2.2
Total production PCCP by manufacturer.....................................................................76
2.3
Total production PCCP- ECP compared to LCP ............................................................77
2.4
Total ECP production by manufacturer ..........................................................................78
2.5
Total LCP production by manufacturer ..........................................................................79
2.6
Minimum cylinder thickness PCCP................................................................................80
2.7
Minimum wire size PCCP ..............................................................................................80
2.8
Mortar coating PCCP......................................................................................................81
3.1
Lifespan histogram all samples.......................................................................................95
3.2
Lifespan histogram installed pre-1955 ...........................................................................96
3.3
Lifespan histogram installed 1955-1963.........................................................................97
3.4
3.5
3.6
Lifespan histogram installed 1972-1978.......................................................................100
3.7
Lifespan histogram installed 1972-1978: Interpace......................................................101
3.8
Lifespan histogram installed 1972-1978: unknown and non-Interpace........................102
3.9
3.10
3.11
Failures by installation date histogram Category 1, 1942-2006 ...................................105
3.12
Failures by installation date with production histogram Category 1, 1942-2006.........106
3.13
Failures by installation date histogram Categories 2 and 3, 1942-2006.......................107
ix
3.14
Failures by installation date with production histogram Categories 2 and 3, 19422006 .......................................................................................................................108
3.15
Normalized failures by installation date histogram Category 1, 1942-2006 ................109
3.16
Normalized failures by installation date histogram Categories 2 and 3,

1942-2006 ............................................................................................................110
3.17
Failure rates by time period .............................................................................................111
3.18
Failure rates by wire class histogram...............................................................................112
3.19
Failure rates by pipe type by year installed histogram Category 1, 1942-2006 ...........113
3.20
Failure rates by pipe type by year installed histogram Categories 2 and 3,

1942-2006 ............................................................................................................114
5.1
Lifespan ogive all samples...............................................................................................138
5.2
Lifespan ogive installed pre-1955....................................................................................139
5.3
Lifespan ogive installed 1955-1963.................................................................................140
5.4
5.5
5.6
5.7
Lifespan ogive installed 1972-1978: Interpace................................................................144
5.8
Lifespan ogive installed 1972-1978: unknown and non-Interpace..................................145
5.9
5.10
5.11
Probability Weibull: all samples (Category 1) .............................................................148
5.12
Probability Weibull: all samples (Categories 1 and 2) .................................................149
5.13
Probability Weibull: all samples (Category 1 and Categories 2 and 3)........................150
5.14
Probability Weibull: all samples (Category 3) .............................................................151
5.15
Probability Weibull: installed pre-1955 (Category 1) ..................................................152
5.16
Probability Weibull: installed pre-1955 (Cateogries 1 and 2) ......................................153
x
5.17
Probability Weibull: installed 1955-63 (Category 1) ...................................................154
5.18
Probability Weibull: installed 1955-63 (Categories 1 and 2) .......................................155
5.19
Probability Weibull: installed 1955-63 (Category 1 and Categories 2 and 3)..............156
5.20
5.21
5.22
Probability Weibull: installed 1964-67 (Category 1 and 2)..........................................159
5.23
5.24
5.25
5.26
5.27
5.28
5.29
5.30
5.31
5.32
Probability Weibull: installed 1992-2006 (Category 1) ...............................................169
5.33
Probability Weibull: installed 1992-2006 (Category 3) ...............................................170
5.34
Weibull-predicted 10-year failure probabilities by time period ......................................171
5.35
Weibull-predicted 100-year failure probabilities by time period ....................................172
6.1
Category 1 PCCPfailures by manufacturing era..............................................................176
6.2
PCCP failures by manufacturing era................................................................................176
xi
FOREWORD
The Awwa Research Foundation is a nonprofit corporation that is dedicated to the
implementation of a research effort to help utilities respond to regulatory requirements and
traditional high-priority concerns of the industry. The research agenda is developed through a
process of consultation with subscribers and drinking water professionals. Under the umbrella of
a Strategic Research Plan, the Research Advisory Council prioritizes the suggested projects
based upon current and future needs, applicability, and past work; the recommendations are
forwarded to the Board of Trustees for final selection. The foundation also sponsors research
projects through the unsolicited proposal process; the Collaborative Research, Research
Applications, and Tailored Collaboration programs; and various joint research efforts with
organizations such as the U.S. Environmental Protection Agency, the U.S. Bureau of
Reclamation (USBR), and the Association of California Water Agencies.
This publication is a result of one of these sponsored studies, and it is hoped that its
findings will be applied in communities throughout the world. The following report serves not
only as a means of communicating the results of the water industrys centralized research
program but also as a tool to enlist the further support of nonmember utilities and individuals.
Projects are managed closely from their inception to the final report by the foundations
staff and large cadre of volunteers who willingly contribute their time and expertise. The
foundation serves a planning and management function and awards contracts to other institutions
such as water utilities, universities, and engineering firms. The funding for this research effort
comes primarily from the Subscription Program, through which water utilities subscribe to the
research program and make an annual payment proportionate to the volume of water they deliver
and consultants and manufacturers subscribe based on their annual billings. The program offers a
cost-effective and fair method for funding research in the public interest.
A broad spectrum of water supply issues is addressed by the foundations research
agenda: resources, treatment and operations, distribution and storage, water quality and analysis,
toxicology, economics, and management. The ultimate purpose of the coordinated effort is to
assist water suppliers to provide the highest possible quality of water economically and reliably.
The true benefits are realized when the results are implemented at the utility level. The
foundations trustees are pleased to offer this publication as a contribution toward that end.
David E. Rager
Chair, Board of Trustees
Robert C. Renner, P.E.

Executive Director
xiii
ACKNOWLEDGMENTS
The authors of this report are indebted to the following water utilities and organizations
for their cooperation and participation in this project:
Ameron International, Aurora Water, California Department of Water Resources,
Calleguas Municipal Water District, Central Arizona Project, City of Aurora, City of
Houston, City of Phoenix, City of San Francisco, Colorado Springs Utilities, Dallas
Water Utilities, Denver Water, Greater Lawrence Sanitary District, Hanson Pipe &
Precast, Louisville Water Company, Metropolitan Water District of Southern California,
New Jersey American Water Co., Openaka Corporation, Pressure Pipe Inspection Co.,
Price Brothers, Pure Technologies US Inc., San Diego County Water Authority, Santa
Clara Valley Water District, So. Central Connecticut Regional Water Authority, Tampa
Bay Water, Tarrant Regional Water District, Tucson Water, U.S. Bureau of Reclamation.
The authors of this report acknowledge the assistance provided by the Project Advisory
Committee:
John H. Bambei, Jr., Chief of Engineering, Denver Water
David Marshall, Engineering Services Manager, Tarrant Regional Water District
Jon Kennedy, Senior Manager of Engineering and Projects, Tampa Bay Water
The AwwaRF Project Manager was Jian Zhang.
The authors of this report are indebted to the following individuals whose voluntary
assistance made the report possible:
Mark L. Bushyeager, Jon Kennedy, Ted Norris, Matt S. Turney, John Galleher, Jr.,
Louise Dion, Randy Randolph, Darrin Francom, Britt Klein, Dana Wright, Brian
Mergalas, Mark Holley, Jonathan Chow, Joe Smith, Larry Catalano, John Marshall,
Randall Payton, Richard E. Weare, Brandy A. Kelso, Daryl A. Little, Steve Fiori, Bob
Beamer, Susan Mulligan, Tim Kennelly, Ron Sanchez, Bret Russell, Dana Wright, David
M. Hughes, Larry Wong, Scot Foreman, James Keith, James Davenport, Bob Card, Sam
Arnaout, Henry Bardakjian, Pamela J. Harren, Armand Tremblay, Steve Rupar, Paul
Acosta, Kim Linton, Tedd R. Calhoun, Alex Margevicius, Roger C. Turcotte, Ed
Padewski, Jay Rechsteiner, Joe Zaccarro, and Iraj Karjoo. Thanks also to Robert
Carpenter of Oildom Publishing Co., who provided a meeting room for the utility
workshop at the Underground Construction Technology Conference, on January 31,
2007.
Finally, the formatting and preparation of the report was by Gloria Dierking.
xv
EXECUTIVE SUMMARY
BACKGROUND
The report of water pipelines failing can be sensational, particularly where the resulting
flood damage provides spectacular footage for the 10 oclock news. These failures, some of
which are prestressed concrete cylinder pipe (PCCP), also cost significant sums for repair and
eventual pipeline replacement. These failures can wash out parallel sanitary sewers, which
present a public health problem due to the possible contamination of the drinking water supply,
and can also destroy private property (Henry, Miller, and Do 2005; Ortega, Henry, and
Hovsepian 2005). Corroded or damaged PCCP pipelines have explosively failed during rapid
changes in pipeline operation.
PCCP tends to be of large diameter, making failures of this type of pipe relatively
catastrophic and costly. To date, most research has been focused on PCCP inspection
technologies and performance prediction in order to minimize the risk to utilities from failures.
The performance of PCCP has been an item of interest to water utilities for a long time. Fifteen
years ago, Awwa Research Foundation (AwwaRF) and the Bureau of Reclamation initiated a
study, Performance of Prestressed Concrete Pipe that was unfortunately never published
(summary statistics were separately published by the Bureau of Reclamation). This study is
intended to disseminate some of the answers that were to be provided by that study and to
provide more complete data useful for the prediction of PCCP service life.
RESEARCH OBJECTIVES
This report presents data collected from this study, as well as expands upon previous
studies on PCCP failures. The project began with a survey of a selected group of water utilities
with a fair amount of PCCP in their systems, selected carefully to represent the range of
conditions and variables that affect PCCP performance across North America. The survey
results, combined with a database of 592 PCCP failures, were statistically projected over the
installed PCCP base of North America and provide a surprising indicator of failures and failure
rate of PCCP.
A timeline was developed to determine when PCCP was made, when it failed, and what
events (such as changes to design and material standards) may have affected its performance.
This timeline was statistically analyzed against the failure data.
SIGNIFICANCE TO WATER UTILITIES
PCCP users know that they have a potential problem. Every pipe has a story to tell.
Central to the cause of problems in waterworks are incomplete data, which would otherwise
allow rational decision-making. The problem is that the data are dispersed and diffuse. Thus, a
primary goal of this study was to collect the data and put it into place at one time.
xvii
APPROACH
A survey and workshop were conducted, including water utilities with a fair amount of PCCP
in their systems. The objective was to make this a utility/user-driven project, utilizing the
following approach:
Define and find the failures, however and wherever possible.

Solicit, collect, and review all possible information about PCCP design, operation,
and failures.
Scrutinize and summarize what is known.
Understand and state what is not known.
Report what was concluded by others.
Take a critical look at the conclusions in light of the increasing knowledge base.
Use statistical tests to determine important factors and begin predictive modeling.
CAUSES AND MODES OF FAILURE

Similar to the human population, each individual pipe section has a birth (manufacturing)
to death cycle that is affected by its heredity (design, manufacturer, materials, etc.), birth defects
(construction, installation), and lifestyle (operations and maintenance, etc.). For the purposes of
this study, failure has been defined as the loss of use of a pipe section or reduction in confidence
in that pipe section to remain in service, after discovery of a pipe section deficiency. This
includes repair, replacement, or reduction in operating pressure.
Three categories or models of PCCP failure were defined:
1. Catastrophic ruptures and leaks (Category 1)
2. Significant deterioration or structural weakness discerned by inspection (Category 2)
by
Visual, sounding, and accidental discovery
Electronic inspections
3. Loss of service (Category 3)
Time out of service
Full or partial replacement
The reasons cited by PCCP owners for the failures that they experienced are many. They
included:
Rupture or break broken wires found after the failure many causes
Leaking at joints many causes including out-of-roundness of joint and construction
damage
Cracks in core many causes including alkali-silica reactivity of the aggregate
Low quality of core poor concrete strength
H2S (force mains) unlined
Dented cylinder fabrication and construction
Cracks in cylinder welds poor cylinder fit-up
xviii
Low quality of wires not just Type IV

Overwrapping of wire inadequate total prestress
Wire spliced and restressed inadequate total prestress
Low quality of mortar low density, low thickness, and low cement content
High chlorides in soil corrosive/aggressive soil inappropriate for mortar-coated pipe
Inadequate joint restraint pipe moved exposing joint to environment
Construction damage coatings damaged and not repaired
Coating delamination many causes
Hydrogen embrittlement of wire excessive cathodic protection applied to
susceptible wire
Inadequate prestress wires broken and spliced without retensioning resulting in low
core compression
Cantilever (bending or broken back) many causes including poor bedding
Settlement general and at structures
Poor bedding not corresponding to design assumption
Surge unanticipated and above design value
Looped gasket joint fit-up
Wrong pipe class pipe laid out of order
Cracks in joint welds poor/no field inspection
Hydrotest pressure in excess of design pressure
Excess external load greater than design assumption
Missing joint coating
FAILURE DATABASE
The failure database has a total of 592 independent entries representing a diverse
collection of Category 1, 2, and 3 failures across all data partitions and across 35 states and the
District of Columbia.
To be included in the database, every entry represented one pipeline that had recorded at
least one failure and a corresponding known location. No other lack of information precluded the
entry from the database.
Of the 592 independent entries, the database includes 435 Category 1 failures, or 61.1
percent of the entries. There were 35,662 Categories 2 and 3 failures, or 45.3 percent of the
entries. Thirty-eight of the entries, or 6.5 percent, specified both a Category 1 and a Categories 2
and 3 failure. Of all the entries, 98.3 percent of the entries had a diameter specified, 97.8 percent
had a pipe type (embedded, lined, etc.) specified or assumed based on the diameter, and 92.9
percent had a wire type (Class I, II, III, or IV) specified or assumed based on the installation
year. In totality, 95.9 percent had the installation date specified and 92.0 percent had a fail date
specified. With these, a total of 85.6 percent of the entries had both the installation date and the
fail date, such that a pipe age could be calculated.
Breaking the database into the partitions in a meaningful manner required the pipeline
age. This left sample populations for further analysis grouped by ranges of dates of different
versions of the AWWA PCCP standards as previously described:
xix
DATA PARTITIONS
PCCP has had a long and diverse history with many changes in standards and materials.
To account for these changes and any impact they might have had on failure rates, the failure
data was analyzed in groups of years that shared similar standards and materials. These groups
were:
Pre-1955 (presumably as manufactured to AWWA 7B.2-T [1949])

195563 (presumably as manufactured to AWWA C301-55 or C301-58)
196467 (presumably as manufactured to AWWA C301-64)
19922007 (presumably as manufactured to AWWA C301-92, although the Principal
Investigators are aware of at least one project specified to AWWA C301-84 in 1994)
HISTOGRAMS AND SIMPLE STATISTICAL ANALYSIS

Using the sample populations, lifespan histograms were produced. When comparing all
failures, it was concluded that the mode, or value that occurs most frequently, of Category 1
failures occurs in 6 to 10 years, while Categories 2 and 3 failures peak at the 26- to 30-year
group. The Categories 2 and 3 failure distribution generally lags the Category 1 failure
distribution by 20 years. The means were 13.95 and 16.75 years for Category 1 and Categories 2
and 3 failures, respectively.
With a total of 4,979,837 pipe produced between 1940 and 2006, the average failure rates
for Category 1 and Categories 2 and 3 were 7.89 x 10-5 and 4.98 x 10-3 failures per pipe
produced, respectively. This indicates statistically that within 50 years of being installed, one
rupture and 66 other failures occurred for every 13,200 pipes (~50 miles of pipe).
Other histograms examine those failed pipelines that were installed within eras
represented by the date of manufacture grouped by effective date of the AWWA standard. The
most interesting histograms are for PCCP pipe installed in 197278. This timeframe represents
an era when a loophole was used to use a higher wire class, Class IV, while still adhering to the
standard. Over three times as many Category 1 failures occurred from this time period than from
the other manufacturing eras.
Another histogram examines failures as a function of wire class. It was observed that the
highest frequency of Category 1 failures occurred with Class IV wire. The second highest
frequency of failures occurred with Class II wire. For Categories 2 and 3 failures, the highest
occurrence of failures also occurred with Class IV wire. The second highest frequency of failures
occurred with Class III wire. The least number of Categories 2 and 3 failures were reported for
Class II wire.
PREDICTING SERVICE LIFE
Histograms and simple statistical analyses examined failures that had already occurred.
Histograms visualize the existing data. The next progression was to utilize these failure data to
forecast future occurrences by cumulative frequency distributions, expressed graphically in a line
xx
graph called an ogive. Ogive uses historical data to make predictions based only on what failed
(no credit given for the population that is still in service). The ogive plots cumulative frequency
versus the upper boundary of a class. The cumulative frequency of PCCP failures were plotted
for each sample population as a function of age.
In order to predict future failure rates, Weibull probability distributions were developed
for the sample populations. Weibull gives credit for population still in service and allows
prediction of failure probability as a function of age. Weibulls are expressed graphically as
unreliability, or probability of failure, versus time on a log-log grapha graph with logarithmic
scales on both axes. Then probabilities were determined for select populations for failures 10 and
100 years of age.
ASSESSMENT MATRICES
A full-day workshop was held with participating utilities on January 31, 2007 at the
Underground Technology Conference, Houston, Texas. Representatives of 15 utilities
participated in the review of the project goals, definitions, and the draft utility assessment
protocol. Those participating represent approximately 6 percent of the PCCP installed (by linear
feet).
In the workshop, the participating utilities were asked, in their opinions, what were the
most important factors related to PCCP failures experienced within their systems. Their
responses were:
Owners risk factors (for material): Wire class and size (No. 1), manufacturer (No. 2),
slurry under wire (No. 3). Also cited were lining cast vs. sprayed, cylinder thickness,
age or date of manufacture (birth), presence of electrical bonding, quality of external
coating, plant vs. site manufactured pipe, and core quality.
Owners inspection and installation factors: inspection and records (No. 1), contractor
(No. 2), pressure class, quality of backfill, handling/transportation, joint
mortaring/diapering, and electrical continuity. Also cited was the presence of physical
preloading (or tension in the wire) actually present in the wire.
Environmental/operations factors: Surge (No. 1), soil (No. 2), cathodic
protection/interference, fluctuating groundwater, surcharging (right-of-way)
management, appurtenance maintenance, ground movement, internal corrosion of
core, and underground to aboveground transition (aerials).
Maintenance factors: Postconstruction inspection and monitoring (No. 1), data
analysis/study, life extension (rehabilitation/repair), design reanalysis.
The participants stressed that in many instances, although it was clear that many factors
influence the longevity of PCCP, there is limited information available for many pipelines. Poor
recordkeeping is the norm, and reconstructing the information even when available is timeconsuming. The participants requested that a two-tier approach be developed, one quickly
determined assessment, and another more extensive initial assessment based upon acquisition
and review of available data. The intent is to be able to determine what, if anything, needs to be
evaluated for any particular pipeline constructed of PCCP. The methodology is equally
applicable to lined-cylinder type PCCP (LC-PCCP) and embedded-cylinder type PCCP
(EC-PCCP).
xxi
The short form assessment matrix that was developed addresses the top five factors that
are both likely to be known about a PCCP pipeline and which have a significant effect on the
pipeline. The longer form assessment matrix addresses many of the factors that are both likely to
be known (or can be discovered with examination of records) about a PCCP pipeline and which
have a significant effect on the pipeline.
The final assessment forms were utilized to evaluate 22 pipelines. Examining the scores
of the long-form assessments it became clear that those that scored below zero had multiple
problems. What was most curious about those with high scores that still experienced failures was
that the common thread was excessive surge pressures. Examination of the contract documents
of those also revealed little evidence of design consideration for the surge or the hydrostatic test
as design conditions. Those two items are red flags waving. Clearly, a PCCP pipeline can have
one or two negatives against it and fail, for the weight of the positives may not override the
weaknesses in the system.
The checklists are thus not intended to be a definitive failure predictor, rather, part of a
multiphased approach. The intent is to provide a tool to the PCCP-owning utility to do a
preliminary self-assessment of its pipelines (using the short form) followed if indicated by a
more refined self-assessment. It was not uncommon for utilities to include with their assessments
comments similar to One thing this showed us is that we don't have nearly as much info as we
should. Carefully completed, it is intended that the assessment checklists, both long and short
form, will be useful as pipe inspection and maintenance prioritization tools and possible input to
risk assessment of PCCP pipelines.
CONCLUSIONS
The initial design basis for manufacture of PCCP appeared to be conservative, and as
experience was gained and competitiveness with other pipe materials increased, changes were
made in the standard to reduce the unit cost of manufacture. Those changes tended to increase
the stress level in the pipe at working pressures and reduced the margin for error. The result was
a significantly increased rate of failure for pipe installed between 1971 and 1979. Fully 50
percent of the catastrophic leaks and breaks recorded were manufactured or installed between
those years.
The trend toward reduced conservatism of the product through revisions in the standard
began to reverse course in 1984 with the issuance of AWWA C301-84. That year saw the
allowable additions of fly ash and other pozzolans in an attempt to increase the density of the
concrete coating and core, the incorporation of ASTM C33 for concrete and mortar aggregate
requirements, the slurry placement under the wire, and the minimum coating thickness increased
to 3/4 inch. Significant revisions to the standard in 1992 and adoption of the very detailed design
standard C304-92 appear to have resulted in much improved performance of as-installed PCCP.
The statistics are summarized in the following table:
xxii
Summary statistics
xxiii
CHAPTER 1
INTRODUCTION
LITERATURE SEARCH
Research to Identify PCCP with High Likelihood of Failure
Early failures of PCCP pipelines prompted visual inspections of many miles of pipelines.
Those inspections often revealed cracks on the interior, leading to further investigations. Interior
cracks sometimes indicated disbondment of the concrete core from the steel cylinder, and it was
soon discovered that tapping on the interior (called sounding) was an effective means for
identifying those areas. These internal inspections were often supplemented with ultrasonic
examination to infer the condition of the core (Lewis and Fisk 2005). An AwwaRF study
(Jackson, Pitt, and Skabo 1992) reviewed the available nondestructive evaluation technology for
waterlines, some of which are applicable to PCCP. But these do not provide any significant
indication of the structural integrity of the prestressing wires.
An alternative inspection methodology was made commercially available in 1997, by
utilizing the wires within the pipe as a radio-frequency measurable coil antenna (Mergalas,
Atherton, and Kong 2001b). Because that technique promised to identify the number and
location of wire breaks in each pipe length inspected, the PCCP owning utilities were interested
in its development to the extent that AwwaRF funded a study (Mergalas and Kong 2001a).
Acoustic monitoring of in-service EC-PCCP pipelines has also been utilized to identify
actively breaking wires (Diaz, Campbell, and Holley 2005; Worthington and DiMarco 1996).
Another real-time technique reported to be successful is inductive scan imaging (Almughery et
al. 2005), a technique that to date has been used only on the exterior of PCCP. Methods
developed to determine the number of wire breaks in EC-PCCP have recently been extended to
LC-PCCP (Kong and Mergalas 2005; Mergalas, Kong, and Balliew 2005). These methods have
widely been promoted to determine the condition of PCCP pipelines, yet their accuracy has not
been as yet demonstrated sufficient to rely entirely thereon (Galleher, Bell, and Romer 2005;
Bambei and Lewis 2005; Parks, Drager, and Ojdrovic 2001). Water utilities, desperate for
concrete answers, continue to fund research (Bengtsson et al. 2005).
It is clear that the time to more actively manage the remaining service life of PCCP
pipelines is at hand. Evaluation and management of PCCP pipelines by utilities has been
approached on the basis of evaluating existing information (Bichler 2005), and a risk-based
approach has been presented (Romer and Bell 2004). A typical PCCP pipeline assessment
approach begins with data collection and analysis (Bell, Kendall, and Mulligan 2001):
Data collection, design, and shop drawings

Complete system surge analysis
System operation modifications
Corrosion survey and alignment corrosivity analysis
Nondestructive investigations
Detailed structural integrity evaluation
1
Crucial data for analyses are often lacking, even at a utility with comprehensive records
(Galleher et al. 2001). Models have been developed to estimate failure risk using fuzzy Markov
techniques (Marshall et al. 2005; Kleiner, Sadiq, and Rajani 2004), when data are scarce or
unreliable. It is not clear that the additional effort provides any greater level of confidence to the
utility, because so little data generally exist.
Remote field-eddy current inspection was initially reviewed in an AwwaRF study
(Jackson, Pitt, and Skabo 1992). This type of inspection, however, is only a component of risk
management (Mergalas et al. 2001c).
AwwaRF has funded a workshop (Lillie et al. 2004), which summarized the devices
available to utilities for inspection of water transmission mains. In that study, the economics of
condition assessment were evaluated against the deferral of capital (replacement of the pipeline)
and operating expenses (repair/rehabilitation). An equally recent AwwaRF study (Reed,
Robinson, and Smart 2004) addressed means of continuous monitoring of the structural capacity
of transmission mains.
Perhaps the most effective, on an individual pipe basis, is analysis of the structural
integrity of the pipe, using the best available data. Those analyses can be as complicated as
finite-element models (Lofti, Oesterle, and Roller 2005, Diab and Bonierbale 2001) or analysis
using the best-available analytic methods at the time the pipeline in question was originally
designed (Lewis and Fisk 2005). Some finite-element models possess a complexity far in excess
of the original pipe design (Gomez et al. 2004).
Alleviating PCCP Failures
Initial steps beyond inspection and analyses at alleviating failure of PCCP included
application of cathodic protection to the prestressing wires (Zarghamee, Ojdrovic, Fongemie
1998), assuming that cathodic protection could reduce the rate of corrosion for corrosion-based
failures. Uncontrolled or unidentified stray currents can cause rapid deterioration of PCCP
pipelines due to hydrogen embrittlement (HE). Even cathodic protection as applied to PCCP in
an effort to control corrosion has been attributed as the cause of failures (Marshall 1998).
Methodologies to identify sources of and isolation of pipelines from stray currents and electric
utility grounding issues were identified in two AwwaRF studies (Duranceau, Schiff, and Bell
1996; Romer and Bell 2004). Others have proposed pulsed cathodic protection (Doniguian, Kips,
and Barnes 1998).
Clearly, there is a significant interest in the preservation and the extension of useful life
of water mains and, in particular, PCCP. Utilities have taken steps in advance of failure to reline
(Fiori, Kendall, and Mulligan 2001), utilizing carbon fiber (Moncreif et al. 2001) and steel
cylinders (Suydam et al. 2001) and replacement with new steel pipe subsequent to failure.
However, even these relined pipelines have been known to fail (McCaffrey, Kendall, and
Mulligan 2006).
Cost of Failures
An accurate method of estimating the tangible and intangible cost of transmission and
distribution water mains failures was introduced in an AwwaRF study (Cromwell et al. 2002). In
that study, electric, transportation, natural gas, and emergency planning industry methods for
estimating customer outage costs were extrapolated to the waterworks industry.
2
The cost of replacement of the nations inventory of PCCP has been estimated at over
$40 billion (Megalas 1998). The cost of maintenance of a PCCP pipeline can also be high
(Marshall, Meza, and Swinnea 2002). AwwaRF has funded studies that focus on the
rehabilitation of distribution mains (Deb et al. 1990; ODay et al. 1986) wherein the cost of
rehabilitation of small-diameter mains approaches the cost of new pipelines. The economics has
not changed significantly in the 16 years since publication of those reports.
Relining PCCP in-place has been generally completed with steel cylinders, starting with
the Bureau of Reclamations Jordan Aqueduct in 1984. Many large aqueducts have been
rehabilitated, at great cost, in this manner (Stine and Stift 1998, Khondiker and Mitchell 1998)
with sometimes significant reduction of flow capacity (and as noted previously, at least one
catastrophic failure).
Survey of Utilities to Supplement Existing Database
Probably more important than simply identifying PCCP leaks and breaks was to establish
if there are patterns of leaks and breaks in PCCP. The Water:\Stats database contains a
tremendous amount of data, but because it lacks information regarding the age, wire type, and
manufacturing characteristics of PCCP, it is not very useful for this study. A more useful set of
data was compiled for AwwaRF (von Fay and Peabody 1994). The PCCP failure rate at that time
was not significantly greater than other pipeline materials, based upon survey results from 114
utilities.
Table 1.1 includes all the PCCP failure data collected since that unpublished study. It
includes information on 583 PCCP failures but omits the cases where data are missing or
incomplete. Figure 1.1 shows the geographic distribution of the lined-cylinder type failures, and
Figure 1.2 shows the geographic distribution of the reported failures of embedded-cylinder type
PCCP. These data represent the most complete body of PCCP failures to the year 2006 extant.
The sources of the data were:
Openakas database
Records accumulated by the Principal Investigators
Previous reports
Published papers
Newspaper and trade press
Manufacturers literature
Interviews
Unpublished information (from participating utilities)
Predicting Failures
The inference of future performance of PCCP based upon analysis of failure rates has
been addressed from an industry perspective (Prosser 1996) and by comparison with other
pipeline materials (von Fay and Peabody 1994). The determination of the remaining useful life
of pipelines, not just PCCP pipelines, has been reported utilizing probabilistic or statistical
methods to estimate a survival function on the basis of past behavior (Nelson 2005). Assignment
of a pipe criticality index included web-based data in a very large PCCP system (Essamin et al.
2005) for EC-PCCP. Artificial neural networks (Najafi and Kulandaivel 2005) have been utilized
3
to predict behavior of systems based upon past behavior. Because of the high numbers of PCCP
pipelines manufactured by Interpace and with Class IV wire, one utility used linear extrapolation
to estimate time of failure (Bradish, Cronin, and Lewis 1995).
The remaining service life has even been estimated on its actuarial value (Baik Shik,
Abraham, and Gipson 2004), although it is understood that utilities would like to believe that
their underground assets have an indefinite useful life. The mathematics proposed (Kleiner,
Rajani, and Sadiq 2005) for managing the risk inherent with PCCP pipes may be significantly
daunting to management, whose eyes roll at the thought of manipulating fuzzy-based Markov
techniques, even when aided by computer programs. What is necessary is a risk assessment
system whose results are readily understandable by water utility management.
The Principal Investigators struggled with the comparison of the effect of PCCP pipeline
failures by valuation. The Principal Investigators do not know the sizes for much of the
production (only LC vs. EC), so comparison by inch-diameter is not possible. The Principal
Investigators do not know how much was the selling price, so the Principal Investigators do not
know the relative worth of each.
And the Principal Investigators really do not know if mostly EC-PCCP or LC-PCCP
fails, only what has been well publicized, that EC-PCCP failures are big enough to be publicized
or subject to lawsuits.
4
Table 1.1
Reported failures of prestressed concrete cylinder pipe
ID
Location
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
1955
1980
25
ALTOONA, PA
24
LCP
19
ARLINGTON, TX
20
LCP
534
ARLINGTON, VA
60
ECP
1972
1972-78
2007
RUPTURED
RUPTURED
IV
105
REPLACEMENT
Comment by investigator
Replacement of approximately 2100 linear feet of

PCCP (project 7000) -BC added manufacturer and
install date per wire class (C301-72, A648,
Interpace practice)
535
ARLINGTON, VA
48
1972-78
2007
IV
525
REDUNDANT LINE
Creation of redundant line prior to inspection and

rehabilitation -BC added manufacturer and install
143
Failure/condition
date per wire class (C301-72, A648, Interpace

practice)
404
ATLANTA, GA
1969
II
RUPTURED
323
AUGUSTA, GA
42
LCP
1974
1989
15
IV
342
AUGUSTA, GA
42
LCP
1974
1991
17
366
AUGUSTA, GA
42
LCP
1974
1995
391
AUGUSTA, GA
42
LCP
1974
1998
RUPTURED
IV
RUPTURED
21
IV
RUPTURED
AT SPIGOT
24
IV
INCIPIENT
PROBABLE SURGE DAMAGE
CORROSION
FAILURE;
INTERNAL
INSPECTION
394
AUGUSTA, GA
42
LCP
1974
1999
25
IV
RUPTURED
EXCAVATION DAMAGE AND ENVIRON.

CORROSION
(continued)
Table 1.1 continued

ID
533
Location
AURORA, CO
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
54
ECP
1980
2000
20
III
Failure/condition
REPAIR
Repair severely distressed pipes 959 (52 wire

breaks) and 958 after PPIC inspection -BC added
wire class/size per install year (C301-79, A648)
65
BALTIMORE, MD
108
ECP
1963
1977
14
LEAKING
APPARENT DAMAGE DURING
90
BALTIMORE, MD
72
ECP
1972
1978
IV
LEAKING
IN TIMBER CRADLE; LINE SHIFTING

LATERALLY
INSTALLATION
113
BALTIMORE, MD
30
LCP
1950
1979
29
144
BALTIMORE, MD
30
LCP
1966
1980
14
II
RUPTURED
SEWER FORCE MAIN; H2S, H2SO4
RUPTURED
COATING DAMAGED TOP OF PIPE; WIRES

CORRODED
207
BALTIMORE, MD
72
ECP
1975
1982
IV
208
BALTIMORE, MD
30
LCP
1980
1982
IV
297
BALTIMORE, MD
36
350
BALTIMORE, MD
72
ECP
1975
1992
17
IV
351
BALTIMORE, MD
72
ECP
1975
1992
17
IV
558
BALTIMORE, MD
72
LCP DW
ECP
1985
2002
LEAKING
EXCESSIVE JOINT OPENING; MOVEMENT
LEAKING
CANTILEVER BENDING FAILURE
RUPTURED
RUPTURED
LINING DISTRESS, INTERNAL INSPECTION
RUPTURED,
20-foot ruptured segment, cracking in two adjacent
REPAIRED
pieces repaired
564
BALTIMORE, MD
36
LCP
2004
RUPTURE
Break March 17, 2004
565
BALTIMORE, MD
36
LCP
2004
RUPTURE
Break January 17, 2004
566
BALTIMORE, MD
36
LCP
2004
RUPTURE
Break March 26, 2004
484
BALTIMORE, MD
2005
54
ECP
228
(continued)
Table 1.1 continued

ID
Location
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
28
BATON ROUGE, LA
48
LCP
1963
1973
10
556
BEULAH, ND
42
ECP,
1979
2005
26
III
2,348
Failure/condition
RUPTURED
three pipes total - sewage force main
REPLACEMENT
Replace with 42 cement-lined and tape-coated
LCP
steel pipe -BC added install date per BOYLE data,

added wire class & size per install date (C301-79,
66
BOCA RATON, FL
24
LCP
1971
1977
II
BOSTON,MA
24
LCP
1949
1957
1/8
209
BRANDON SHORES,
78
ECP
1980
1982
IV
102
ECP
1980
1986
102
ECP
1980
1990
102
ECP
1980
102
ECP
102
LEAKING
SEWER FORCE MAIN H2S TAIL OFF
RUPTURED
MBOT
RUPTURED
SETTLEMENT AND GRADE E STEEL
IV
RUPTURED
10
IV
INCIPIENT FAILURE; INTERNAL INSPECTION
1991
11
IV
1980
1992
12
IV
13
ECP
1980
1993
13
IV
102
ECP
1980
1995
15
IV
102
ECP
1980
1996
16
IV
MD
A648)
301
BRANDON SHORES,
MD
330
BRANDON SHORES,
MD
343
BRANDON SHORES,
MD
352
BRANDON SHORES,
MD
359
BRANDON SHORES,
MD
367
BRANDON SHORES,
MD
375
BRANDON SHORES,
MD
(continued)
Table 1.1 continued

ID
382
Location
BRANDON SHORES,
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
Failure/condition
(in.)
type
date
date
yrs.
class
size
ruptures
failures
102
ECP
1980
1997
17
IV
78
ECP
1980
1997
17
IV
102
ECP
1980
1998
18
IV
102
ECP
1980
1999
19
IV
17
102
ECP
1980
1999
19
IV
MD
383
BRANDON SHORES,
MD
BRANDON SHORES,
MD
395
BRANDON SHORES,
MD
396
BRANDON SHORES,
RUPTURED
MD
392
ENVIRONMENTAL CORROSION AND

SURGE EVENTS
256
BRECKENRIDGE,MN
16
LCP
1955
1983
28
BRIDGEPORT, CT
24
LCP
1954
1968
14
LEAKING
LOOPED GASKET
LEAKING
COATING & PIPE DAMAGED DURING

INSTALLATION
210
BROADVIEW, IL
24
LCP
1966
1982
16
II
12
BROCKTON, MA
30
LCP
1966
1971
II
1
1
LEAKING
EXCESS JOINT OPENING
RUPTURED
LOOPED GASKET ERODED MORTAR AND

STEEL
173
BUCKS CO. PA
48
ECP
1969
1981
12
II
RUPTURED
APPEARS TO BE BY CONTRACTOR
BARREL, BROKEN
WIRES
33
BURNSVILLE, MN
36
LCP
1970
1974
II
174
BURNSVILLE, MN
36
LCP
1971
1981
10
II
1
1
RUPTURED
DEFECTIVE WIRE
LEAKING
FLEX TIED JOINT SEPERATED; UNSTABLE

BEDDING
211
BURNSVILLE, MN
36
LCP
1968
1982
14
II
BROKEN WIRES
CATHODIC INTERFERENCE
(continued)
Table 1.1 continued

ID
427
Location
CALABASAS, CA
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
54
ECP
1975
II
8-
Failure/condition
14
14 pipe sections with breaks; the maximum
1/4
469
CALLEGUAS, CA
51
ECP
1962
1988
26
II
448
CALLEGUAS, CA
48
ECP
1962
1999
37
II
number of wire breaks in any single section is 15.

1
5
RUPTURE
1988 Leak & Pipe Replacement
Wire breaks: very low
partial RECT/TC
pipe
449
CALLEGUAS, CA
78
ECP
1971
2001
30
II
8, 6,
1/4
Wire breaks: brittle
RECT/TC
wires from impressed

current
number of breaks per
450
CALLEGUAS, CA
78
ECP
1975
2001
26
III
RECT/TC

current
451
CALLEGUAS, CA
66
ECP
1980
2001
21
III
8, 6
RECT/TC

current
447
CALLEGUAS, CA
51
ECP
1962
2002
40
II
Wire breaks: very low
partial RECT/TC
number of breaks per

pipe
470
CALLEGUAS, CA
51
ECP
1962
2002
40
II
536
CALLEGUAS, CA
51
ECP
1962
2005
43
II
165
7
REPLACEMENT
2002 Steel Liner Project
RUPTURE
The steel liner had failed along the longitudinal

seam for approximately 42.7 linear meters (140
linear feet)
468
CALLEGUAS, CA
51
ECP
1962
2007
45
II
RUPTURE
Failure at Yosemite, January 2007
471
CALLEGUAS, CA
51
ECP
1962
2007
45
II
RUPTURE
Failure west of Stearns St., February 2007
(continued)
Table 1.1 continued

ID
Location
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
1979
27
114
CALVERT CITY, KY
24
LCP
1952
428
CASTAIC, CA
201
ECP
1968
II
1
8-
Failure/condition
RUPTURED
PIPELINE UNDER A SALT PILE
Moderately good condition; 3 pipes with 20+
1/4
50
CENTER, ND
30
LCP
1974
1976
IV
breaks
2
LEAKING
COOLING H2O MAKEUP PIPE 1.5 YEARS
466
CENTER, ND
30
LCP
1974
1977
IV
91
CENTER, ND
30
LCP
1976
1978
IV
340
1
REPLACED
6800' replaced in 1977
RUPTURED
SHEARED AT FACE OF CONCRETE THRUST

BLOCK
10
OLD
34
CHESTER, PA
42
LCP
1968
1974
II
LEAKING
COATING DAMAGED DURING

INSTALLATION
212
CHESTER, PA
42
LCP
1968
1982
14
II
RUPTURED
257
CHESTER, PA
66
ECP
1974
1983
IV
RUPTURED
453
CHICAGO, IL
30
LCP
1954
1959
MB
RUPTURED
OT
CHEMICAL PROBLEM .. IN CORN FIELD
Lewiston Pipe (non-cylinder) Possible external

load fromm mfooter
285
CHICAGO, IL
42
LCP
1972
1984
12
IV
RUPTURED
CORROSION
454
CHICAGO, IL
42
LCP
1972
1984
12
IV
RUPTURED
CORROSION
13
CLEVELAND, OH
48
LCP
1964-67
1971
II
RUPTURED
-BC added install date per wire class (C301-64,

A227)
14
CLEVELAND, OH
48
ECP
1956
1971
15
RUPTURED
CHOLRIDE ION CONTAMINATION
40
CLEVELAND, OH
30
LCP
1955
1975
20
RUPTURED
AGGRESSIVE GROUND WATER, CHLORIDE

IONS
213
CLEVELAND, OH
30
LCP
1956
1982
26
LEAKING
CHOLRIDE IONS
(continued)
Table 1.1 continued

ID
Location
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
384
CLEVELAND, OH
36
LCP
1968
1997
29
II
51
CLOQUET, MN
36
LCP
1968
1976
16
II
Failure/condition
RUPTURED
DENTED CYLINDER
LEAKING
CHUNK OF CONCRETE DROPPED AGAINST

COATING
67
CLOQUET, MN
36
LCP
1968
1977
II
LEAKING
COATING APPARENTLY DAMAGED WHEN
68
CLOQUET, MN
36
ECP
1968
1977
II
RUPTURED
PUMP STATION CONTROLS

MALFUNCTIONED
11
INSTALLED
92
CLOQUET, MN
36
LCP
1976
1978
IV
RUPTURED
CORROSION
115
CLOQUET, MN
36
ECP
1968
1979
11
II
RUPTURED
FLEX TIED WELDED JT. FAILED DUE TO

SURGE
214
CLOQUET, MN
36
LCP
1968
1982
14
II
LEAKING
CORRODED WIRES; INSTALLATION

RELATED
215
CLOQUET, MN
36
ECP
1968
1982
14
II
LEAKING; BROKEN
SURGE; UNSTABLE SOIL
WIRES
216
CLOQUET, MN
36
LCP
1968
1982
14
II
LEAKING, BROKEN
CONTRACTOR FAULTY WELDING
WIRES
217
CLOQUET, MN
36
LCP
1968
1982
14
II
BROKEN WIRE
CONTRACTOR DAMAGED; BEDDING OUT

OF SPEC.
COBB CO., GA
30
LCP
1951
1967
16
RUPTURED
COBB CO., GA
30
LCP
1952
1967
15
RUPTURED
20
COBB CO., GA
30
LCP
1951
1972
21
RUPTURED
21
COBB CO., GA
30
LCP
1952
1972
20
RUPTURED
DAMAGED DURING INSTALLATION
WATER HAMMER FOLLOWING POWER

SURGE
(continued)
Table 1.1 continued

ID
Location
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
Failure/condition
12
29
COBB CO., GA
36
ECP
1969
1973
II
RUPTURED
35
COBB CO., GA
36
ECP
1969
1974
II
RUPTURED
36
COBB CO., GA
36
ECP
1969
1974
II
RUPTURED
41
COBB CO., GA
30
LCP
1951
1975
24
RUPTURED
42
COBB CO., GA
30
LCP
1952
1975
23
RUPTURED
52
COBB CO., GA
30
LCP
1951
1976
25
RUPTURED
53
COBB CO., GA
30
LCP
1951
1976
25
RUPTURED
54
COBB CO., GA
30
LCP
1952
1976
24
LEAKING
CORROSION
55
COBB CO., GA
36
ECP
1969
1976
II
BROKEN WIRES
100 FT APART
69
COBB CO., GA
36
ECP
1969
1977
II
BROKEN WIRES
ABOUT 3 MOS. APART; SAME AREA
70
COBB CO., GA
36
ECP
1969
1977
II
RUPTURED
71
COBB CO., GA
36
ECP
1969
1977
II
RUPTURED
72
COBB CO., GA
36
ECP
1969
1977
II
RUPTURED
2ND PIPE FROM PREVIOUS FAILURE
SURGE AND CORROSION
1 DAY AFTER PREVIOUS ONE; MOVED

VERTICALLY
73
COBB CO., GA
36
ECP
1969
1977
II
74
COBB CO., GA
36
ECP
1969
1977
II
93
COBB CO., GA
36
LCP
1966
1978
12
II
94
COBB CO., GA
36
LCP
1967
1978
11
II
95
COBB CO., GA
36
ECP
1969
1978
II
RUPTURED
116
COBB CO., GA
36
ECP
1969
1979
10
II
RUPTURED
2
1
RUPTURED
200 FT FROM PREVIOUS FAILURES
BROKEN WIRES
SURGES
RUPTURED
1
BROKEN WIRES
FOUND UNCOVERED FOR PRESSURE TEST
CORRODED WIRE
(continued)
Table 1.1 continued

ID
117
Location
COBB CO., GA
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
36
ECP
1969
1979
10
II
Failure/condition
RUPTURED
COATING GOUGED AND POSSIBLY

PATCHED
118
COBB CO., GA
36
LCP
1973
1979
IV
1 LEAKING; 2 WITH
CORRODED WIRE
BROKEN WIRES
COBB CO., GA
36
LCP
1974
1979
IV
1 LEAKING; 1 WITH
LOOPED GASKET; POOR COATING
BROKEN WIRE
145
COBB CO., GA
36
ECP
1969
1980
11
II
BROKEN WIRES
RIVER CONN. BALLOONED; CTG. OFF FOR 1

FOOT
146
COBB CO., GA
48
ECP
1973
1980
IV
BROKEN WIRES
13
119
HEALED ALONG CRACKS IN RUPTURED

PIPE
147
COBB CO., GA
48
ECP
1975
1980
IV
175
COBB CO., GA
36
LCP
1969
1981
12
II
176
COBB CO., GA
36
ECP
1970
1981
11
II
177
COBB CO., GA
36
LCP
1973
1981
IV
RUPTURED
BROKEN WIRE
LEAKING
CONTRACTOR DAMAGED COATING
RUPTURED
CORRODED WIRE
RUPTURED
THIN COATING AND WRONG CLASS OF
PIPE
178
179
COBB CO., GA
COBB CO., GA
36
36
LCP
LCP
1973
1973
1981
1981
IV
IV
LEAKING WITH
THIN COATING; CEMENT CONTENT; WIRE
BROKEN WIRES
BRIDGING
LEAKING
THIN CTG.; CEMENT CONTENT; WIRE

BRIDGING
180
COBB CO., GA
36
LCP
1973
1981
IV
LEAKING
THIN CTG.; CEMENT CONTENT; WIRES;

POSS. SURGE
181
COBB CO., GA
48
ECP
1974
1981
IV
RUPTURED
CORRODED WIRE
182
COBB CO., GA
48
ECP
1974
1981
IV
RUPTURED
CORRODED WIRE
(continued)
Table 1.1 continued

ID
Location
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
Failure/condition
183
COBB CO., GA
36
LCP
1975
1981
IV
RUPTURED
WIRE FRACTURE
184
COBB CO., GA
36
LCP
1975
1981
IV
RUPTURED
OUT-OF-ROUND SPIGOT; CRACKED

COATING
14
218
COBB CO., GA
30
LCP
1951
1982
31
RUPTURED
SURGES
219
COBB CO., GA
30
LCP
1951
1982
31
RUPTURED
SURGE
220
COBB CO., GA
36
ECP
1969
1982
13
II
RUPTURED
BAD WELD
221
COBB CO., GA
48
ECP
1973
1982
IV
RUPTURED
MANY FACTORS
258
COBB CO., GA
36
ECP
1970
1983
13
II
RUPTURED
ONE DAY AFTER PREVIOUS RUPTURE
259
COBB CO., GA
36
ECP
1970
1983
13
II
RUPTURED
POSSIBLE HOLLOW COAT
260
COBB CO., GA
36
ECP
1970
1983
13
II
RUPTURED
POSSIBLE HOLLOW COAT
261
COBB CO., GA
36
ECP
1970
1983
13
II
RUPTURED
262
COBB CO., GA
36
LCP
1974
1983
IV
RUPTURED
5TH FAILURE IN THIS AREA
263
COBB CO., GA
48
ECP
1974
1983
IV
RUPTURED
7 TH FAILURE IN THIS AREA
264
COBB CO., GA
48
ECP
1978
1983
IV
RUPTURED
6TH FAILURE IN THIS AREA
265
COLORADO SPRINGS,
36
LCP
1981
1983
IV
RUPTURED
POOR DESIGN
48
LCP
1949
1982
33
RUPTURED
wires cut by trencher 2 weeks prior -BC added
CO.
222
CORAL GABLES, FL
wire class/size per install date

1
CORPUS CHRISTI, TX
24
LCP
1955
455
COSTA MESA, CA
40
LCP
1975
RUPTURED
(continued)
Table 1.1 continued

ID
Location
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
120
DALTON, GA
30
LCP
1953
1979
26
223
DeKALB, GA
48
LCP
1961
1982
21
Failure/condition
RUPTURED
DAMAGED COATING
RUPTURED
CONTRACTOR DAMAGE OR TIGHT JOINT

(1983 per PBC list 4/7/86)
DENVER, CO
66
ECP
1977
1997
20
IV
1/4
RUPTURED
393
DENVER, CO
48
LCP
1972
1998
26
IV
RUPTURED
397
DENVER, CO
48
LCP
1972
1999
27
IV
10
INCIPIENT
CATHODIC PROTECTION/INTERFERENCE
FAILURE;
INTERNAL
INSPECTION
15
385
407
DENVER, CO
36
LCP
1952
42
REPLACED
Most replaced - all but approx 125' remains
408
DENVER, CO
36
LCP
1962
15
REPLACED
Almost all replaced
409
DENVER, CO
48
LCP
1964
90
REPLACED
Replaced
410
DENVER, CO
48
LCP
1974
IV
326
REPLACED
Replaced
411
DENVER, CO
48
LCP
1974
IV
195
REPLACED
Replaced
412
DENVER, CO
48
LCP
1974
IV
281
REPLACED
Replaced
413
DENVER, CO
72
ECP
1974
IV
REPLACED
Some sections of Conduit No. 94 replaced - need

to determine which.
414
DENVER, CO
72
ECP
1974
IV
40
REPLACED

to determine which.
415
DENVER, CO
72
ECP
1974
IV
REPLACED

to determine which.
416
DENVER, CO
72
ECP
1974
IV
REPLACED

to determine which.
(continued)
Table 1.1 continued

ID
417
Location
DENVER, CO
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
60
ECP
1974
IV
Failure/condition
REPLACED

to determine which.
418
DENVER, CO
66
ECP
1977
IV
REPLACED

to determine which.
DENVER, CO
66
ECP
1977
IV
REPLACED

to determine which.
420
DENVER, CO
66
ECP
1977
IV
REPLACED

to determine which.
421
DENVER, CO
66
ECP
1977
IV
REPLACED
16
419

to determine which.
422
DENVER, CO
66
ECP
1977
IV
REPLACED

to determine which.
423
DENVER, CO
66
ECP
1977
IV
REPLACED

to determine which.
424
DENVER, CO
66
ECP
1977
IV
1/4
REPLACED

to determine which.
121
DETROIT, MI
48
ECP
1972
1979
III
224
DETROIT, MI
42
ECP
1972
1982
10
III
148
DULUTH, MN
42
LCP
1971
1980
II
149
DULUTH, MN
42
LCP
1974
1980
IV
185
DULUTH, MN
42
LCP
1975
1981
225
DULUTH, MN
42
LCP
1975
1982
150
DURHAM, NC
30
LCP
1974
1980
LEAKING
LOOPED GASKET
RUPTURED
SLAG INCLUSION ALLOWED WIRE TO FAIL
RUPTURED
RUPTURED
SURGE WHEN SHIFTING PUMPS
IV
RUPTURED
SURGES
IV
RUPTURED
SURGE
IV
RUPTURED
LOOPED GASKET
(continued)
Table 1.1 continued

ID
56
Location
EAST ORANGE, NJ
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
24
LCP
1965
1976
11
II
Failure/condition
LEAKING
LARGE ROCK AGAINST COATING IN BACK

FILL
EAST ORANGE, NJ
24
LCP
1965
1978
13
II
LEAKING
HOLE PUNCHED THROUGH WALL
75
EDISON, NJ
20
LCP
1974
1977
IV
LEAKING
DAMAGED DURING INSTALLATION
76
EDISON, NJ
20
LCP
1974
1977
IV
LEAKING
SEWER FORCE MAIN DAMAGED AT

INSTALLATION
97
EDISON, NJ
20
LCP
1974
1978
IV
LEAKING
DAMAGE TO COATING AFTER

INSTALLATION
17
96
286
EL PASO, TX
60
ECP
1977
1984
III
RUPTURED
287
EL PASO, TX
60
ECP
1977
1984
III
RUPTURED
288
EL PASO, TX
60
ECP
1977
1984
III
BROKEN WIRES,
PBC list 4/7/86 indicates inst. range 1972-78
POROUS COATING; BROKEN WIRES
EXTERNAL
INSPECTION
289
EL PASO, TX
54
ECP
1979
1984
III
317
EL PASO, TX
54
ECP
1979
1988
III
13
RUPTURED
BROKEN WIRES,
EXTERNAL
INSPECTION
344
EL PASO, TX
60
ECP
1976
1991
15
III
403
EL PASO, TX
54
ECP
1979
1992
13
III
1
5
RUPTURED
CYLINDER LEAK
LINING DISTRESS,
INTERNAL
INSPECTION
(continued)
Table 1.1
ID
361
Location
EL PASO, TX
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
54
ECP
1979
1994
15
III
Failure/condition
LINING DISTRESS,
INTERNAL
INSPECTION
EL PASO, TX
527
EL PASO, TX
54
ECP
1979
1995
16
60,
ECP,
1977
2003
26
54,
LCP
1955-63
2005
III
6
6
5
1

FAILURE
42
18
368
526
EL PASO, TX
48
ECP
AWWA C301. -BC added wire class and install

date per wire size (C301-55 or C301-58, A227)
43
ERIE CO, PA
42
ECP
1971
1975
II
LEAKING
WELD FAILURE BETWEEN JOINT RING &

CYLINDER
226
FAIRFAX CITY, VA
22
FAIRFAX CO. VA
24
LCP
1960
1982
22
RUPTURED
LCP
1962
1972
10
RUPTURED
LOOPED GASKET; ERODED MORTAR AND

STEEL
44
FAIRFAX CO. VA
30
LCP
1960
1975
15
RUPTURED
57
FAIRFAX CO. VA
24
LCP
1959
1976
17
RUPTURED
58
FAIRFAX CO. VA
30
LCP
1947
1976
29
77
FAIRFAX CO. VA
30
LCP
1947
1977
30
98
FAIRFAX CO. VA
30
LCP
1947
1978
31
99
FAIRFAX CO. VA
36
LCP
1975
1978
IV
122
FAIRFAX CO. VA
30
LCP
1965
1979
14
II
BROKEN WIRES
IN SECTION OF LINE RELAID IN 1964
RUPTURED
IN SECTION OF LINE RELAID IN 1964
RUPTURED
RUPTURED
RUPTURED
(continued)
Table 1.1
ID
Location
123
FAIRFAX CO. VA
456
FAIRFAX CO. VA
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
36
LCP
1975
1979
IV
LCP
1976
1979
Failure/condition
BROKEN WIRES
RUPTURED
PBC list 4/7/86 indicates failure. range 1979-80.

It's not clear that these are in the above listed.
19
186
FAIRFAX CO. VA
36
LCP
1975
1981
IV
LEAKING
COATING CRACK
187
FAIRFAX CO. VA
30
LCP
1975
1981
IV
LEAKING
BELL BOLT JOINT; GRADE E CYLINDER
227
FAIRFAX CO. VA
36
LCP
1975
1982
IV
RUPTURED
298
FAIRFAX CO. VA
36
LCP
1975
1985
10
IV
RUPTURED
299
FAIRFAX CO. VA
36
LCP
1975
1985
10
IV
RUPTURED
302
FAIRFAX CO. VA
36
LCP
1975
1986
11
IV
RUPTURED
331
FAIRFAX CO. VA
36
LCP
1975
1990
15
IV
RUPTURED
332
FAIRFAX CO. VA
36
LCP
1975
1990
15
IV
RUPTURED
369
FAIRFAX CO. VA
36
LCP
1975
1995
20
IV
RUPTURED
386
FAIRFAX CO. VA
36
LCP
1975
1997
22
IV
RUPTURED
530
FAIRFAX CO. VA
36
LCP
2001
1,000
REPLACEMENT
531
FAIRFAX CO. VA
36
LCP
2002
400
REPLACEMENT
228
FAYETTEVILLE, NC
48
LCP
1968
1982
14
II
RUPTURED
TORN CYLINDER
THRUST PROBLEM; WIRE SPACING TOO

CLOSE 4/7/86 PBC lists 1 yr. newer, 1 year earlier
failure.
387
FAYETTEVILLE, NC
229
GAFFNEY, SC
42
LCP
1969
1997
28
IV
RUPTURED
1982
1982
III
CYLINDER WELD FAILURE
(continued)
Table 1.1
ID
Location
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
GAINESVILLE, FL
36
LCP
1968
1981
13
II
266
GAINESVILLE, FL
48
LCP
1969
1983
14
151
GRAHAM, NC
30
LCP
1974
1980
IV
78
GRAND FORKS, ND
30
LCP
1960
1977
17
230
GRAND FORKS, ND
30
LCP
1959
1982
23
457
GRAND JUNCTION,
24
LCP
1964
1979
II
LEAKING
JOINT RING
LEAKING
RUPTURED
LEAKING IN SEAM WELD
LEAKING
H2S
LEAKING
FAULTY AIR RELEASE VALUE OPERATION
Corrosion -BC added wire class/size per install
CO
20
188
Failure/condition
529
GRAND VALLEY, CO
date (C301-64, A227)

42,
LCP
36,
early
2001
4,884
REPLACEMENT
1960s
The pipeline is presently subject to frequent breaks

due to deteriorated pipe condition, and is
24
unreliable due to its location within geologic

hazards and stream erosion areas.
189
GRANITE CITY, IL
24
LCP
1958
1981
23
RUPTURED
STEEL PLANT - RUPTURED IN AREA OF

JOINT
190
GREENVILLE, SC
30
541
GREENVILLE, SC
48
539
GREENVILLE, SC
42,
LCP
1981
RUPTURED
1976
1988
12
IV
RUPTURES
LCP
1974
1990
16
IV
RUPTURE
LCP
1974
1994
20
IV
RUPTURES
1974
1997
23
IV
48
540
GREENVILLE, SC
42,
48
537
GREENVILLE, SC
48
370
REPLACEMENT
Remove and replace 7,400 feet of 48" PCCP with

48" CIP (contract 85)
(continued)
Table 1.1
ID
538
Location
GREENVILLE, SC
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
Failure/condition
20,
LCP
1973
1997
24
IV
RUPTURES
ECP
1979
1997
18
IV
RUPTURES
24
542
GREENVILLE, SC
72
Failures occurred near creek crossings. Appeared
543
GREENVILLE, SC
72
ECP
1979
1999
20
IV
RUPTURE
Failures occurred near creek crossings. Appeared

to be due to hydrogen embrittlement
544
GREENVILLE, SC
72
ECP
1979
2000
21
IV
2,785
PARALLEL MAIN
Parallel transmission main installed due to

diminished reliablity of existing 72-inch main.
Contracts 87, 91, 93 made redundant 55,700 feet
21
to be due to hydrogen embrittlement
of PCCP
549
GREENVILLE, SC
24
LCP
1973
2001
28
IV
490
PARALLEL MAIN
Parallel transmission main installed due to

diminished reliablity of existing 24-inch main.
Contracts 92 made redundant 9,800 feet of PCCP BC added wire size per install date (C301-72,
A648)
545
GREENVILLE, SC
48
1974
2004
30
IV
685
REPLACEMENT
Remove and replace 13,700 feet of 48" PCCP,

some slipplined (contract 95)
546
GREENVILLE, SC
42
LCP
1974
2005
31
IV
60
REPLACEMENT

42" DIP (contract 99)
547
GREENVILLE, SC
42
LCP
1974
2006
32
IV
275
REPLACEMENT

42" DIP (contract 100)
548
GREENVILLE, SC
36
LCP
1971
III
RUPTURES
BC added wire class/size per year installed (C30164, A227)
100
GWINETT CO, GA
48
LCP
1975
1978
IV
RUPTURED
250 FT DOWN SLOPE FROM EARLIER ONE
(continued)
Table 1.1
ID
101
Location
GWINETT CO, GA
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
48
LCP
1975
1978
IV
Failure/condition
BROKEN WIRES
SURGE - 100FT DROP BETWEEN AIR

VALVES
GWINETT CO, GA
48
LCP
1975
1980
IV
RUPTURED
153
GWINETT CO, GA
48
LCP
1975
1980
IV
RUPTURED
BETWEEN 2 IN 1978
1500 FT FROM EARLIER BREAKS AND
CLOSE TO END
154
GWINETT CO, GA
48
LCP
1975
1980
IV
RUPTURED
2300 FT FROM EARLIER BREAK NEAR END

OF LINE
155
GWINETT CO, GA
48
LCP
1975
1980
IV
RUPTURED
22
152
NEAR PREVIOUS ONE; 2300 FT FROM 1978

RUPTURE
191
GWINETT CO, GA
48
LCP
1975
1981
IV
RUPTURED
OVERWRAPING
192
GWINETT CO, GA
48
LCP
1975
1981
IV
RUPTURED
OVERWRAPPING
193
GWINETT CO, GA
48
LCP
1975
1981
IV
RUPTURED
PIPE OVERWRAPPING
194
GWINETT CO, GA
48
LCP
1975
1981
IV
RUPTURED
OVERWRAPPING
195
GWINETT CO, GA
48
LCP
1980
1981
IV
RUPTURED
DESIGN CONSIDERATIONS
303
GWINETT CO, GA
48
LCP
1976
1986
10
IV
RUPTURED
4/7/86 PBC list indicates 14 failures between 1978

& 1981
307
GWINETT CO, GA
48
LCP
1976
1987
11
IV
318
GWINETT CO, GA
48
LCP
1976
1988
12
IV
RUPTURED
324
GWINETT CO, GA
48
LCP
1976
1989
13
IV
RUPTURED
333
GWINETT CO, GA
48
LCP
1976
1990
14
IV
345
GWINETT CO, GA
48
LCP
1976
1991
15
IV
RUPTURED
RUPTURED
(continued)
Table 1.1
ID
Location
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
Failure/condition
23
353
GWINETT CO, GA
48
LCP
1976
1992
16
IV
362
GWINETT CO, GA
48
LCP
1976
1994
18
IV
RUPTURED
370
GWINETT CO, GA
48
LCP
1976
1995
19
IV
RUPTURED
376
GWINETT CO, GA
48
LCP
1976
1996
20
IV
156
HACKENSACK, NJ
48
LCP
1968
1980
12
II
LEAKING
102
HADDON HEIGHTS, NJ
16
LCP
1964
1978
14
II
LEAKING AT TAP
458
HENDERSON, NC
467
HESPERIA, CA
1976
132
ECP
1971
1988
RUPTURED
1
17
II
8, 6,
THRUST BLOCK AT ELBOW REMOVED
4/7/86 PBC list indicates 1 failure
RUPTURE
High Ground Water, Electrically shorted to water

services and reinforcing steel in "Y" vault.
Corrosion Pipe Failure in 1988, 132in pipe
abandoned in place
103
HINCKLEY, ME
36
LCP
1975
1978
IV
LEAKING
LAID CANTILEVERED ON ROCK LEDGECRUSHED
354
HOPE CREEK, NJ
144
ECP
1976
1992
16
IV
1/4
FOUND DURING
WIRES SPLICED AND RESTRESSED
EXTERNAL
INSPECTION
388
HOPE CREEK, NJ
144
ECP
1976
1997
21
IV
1/4
FOUND DURING
WIRES SPLICED AND RESTRESSED
EXTERNAL
INSPECTION
30
HOT SPRINGS, AK
24
LCP
1968
1973
II
RUPTURED
LAID ON ROCK
45
HOT SPRINGS, AK
24
LCP
1968
1975
II
RUPTURED
LAID ON ROCK
79
HOT SPRINGS, AK
24
LCP
1968
1977
II
RUPTURED
MAY HAVE BEEN WRONG CLASS
(continued)
Table 1.1
ID
551
Location
HOUSTON, TX
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
108
ECP
1972
1990
18
IV
Failure/condition
RUPTURE
"Renovation of three 108-inch PCCP water

pipelines" by Ralph T Rundle, John M Olden.
Hydrogen Embrittlement from CP. -BC added
wire size per install date (C301-72, A648)
HOUSTON, TX
108
ECP
1972
1991
19
IV
RUPTURE
"Renovation of three 108-inch PCCP water

pipelines" by Ralph T Rundle, John M Olden.
Hydrogen Embrittlement from CP. -BC added
wire size per install date (C301-72, A648)
24
552
474
HOUSTON, TX
60
ECP
1976
2002
26
475
HOUSTON, TX
42
LCP
1992
2004
12
III
0.19
99
1,299
prestressed wires failed due to hydrogen embrittlement

corrosion coinciding
sliplined with 36-inch HDPE pipe. Lockwood,
with cracks in the
Andrews & Newnam, Inc. (LAN) assisted in the
mortar linings
failure assessment. -BC added wire class and size

per install date (C301-92, A648)
124
HOWARD CO., MD
36
LCP
1976
1979
IV
LEAKING
MANIFOLD PIPE W/24 OUTLET; TEAR IN

CYLINDER
490
HOWARD CO., MD
30,
LCP
1972-78
1979
IV
36
CATASTROPHIC
-BC added install date, wire size, and
BURST
manufacturer per wire class (C301-72, A648,

Interpace practice)
231
HOWARD CO., MD
36
LCP
1974
1982
IV
RUPTURED
THIN COATING; LEAN COATINGS
232
HOWARD CO., MD
36
LCP
1975
1982
IV
RUPTURED
UNKNOWN
491
HOWARD CO., MD
30,
LCP
1972-78
1982
IV
CATASTROPHIC
BURST
36
Interpace practice)
(continued)
Table 1.1
ID
492
Location
HOWARD CO., MD
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
30,
LCP
1972-78
1982
IV
36
Failure/condition
CATASTROPHIC
BURST

Interpace practice)
HOWARD CO., MD
36
LCP
1974
1988
14
IV
RUPTURED
DELAMINATED BELL COATING; DENTED

CYLINDER
493
HOWARD CO., MD
30,
LCP
1972-78
1988
IV
36
CATASTROPHIC
BURST

Interpace practice)
25
319
325
HOWARD CO., MD
36
LCP
1974
1989
494
HOWARD CO., MD
30,
LCP
1972-78
1989
15
IV
IV
1
1
36
334
HOWARD CO., MD
36
LCP
1974
1990
16
IV
LEAKING
COLLAPSED ON EXCAVATION
LEAK, PIPE
COLLAPSED UPON
EXCAVATION
Interpace practice)
LINING DISTRESS,
LONGITUDINALLY CRACKED LININGS
INTERNAL
INSPECTION
495
HOWARD CO., MD
30,
LCP
1972-78
1990
IV
36
CATASTROPHIC
BURST

Interpace practice)
346
HOWARD CO., MD
36
LCP
1974
1991
496
HOWARD CO., MD
30,
LCP
1972-78
1991
17
IV
IV
1
1
JOINT LEAK
JOINT LEAK
36
LOOPED GASKET
Interpace practice)
355
HOWARD CO., MD
36
LCP
1974
1992
18
IV
RUPTURED
(continued)
Table 1.1
ID
Location
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
IV
360
HOWARD CO., MD
42
LCP
1974
1993
19
489
HOWARD CO., MD
42
LCP
1974
1993
19
497
HOWARD CO., MD
30,
LCP
1972-78
1996
Failure/condition
RUPTURED
2
IV
710
PURE investigated. Under Capital Project W-
REPLACED
8169, the Elkridge Transmission was replaced in

1996. -BC added install date, wire size, and
manufacturer per wire class (C301
488
HOWARD CO., MD
42
LCP
1974
2000
26
Accoustically monitored by PURE, detected

numerous wire breaks between Sta.49+62 and
Sta.50+20. Eighty linear feet PCCP replaced.
26
36
COMPLETELY
488
HOWARD CO., MD
42
LCP
1974
2002
28
385
Howard County's Bureau of Engineering initiated

a new capital project for fiscal year 2002 to repalce
the entire 7,700 linear feet of 42-inch PCCP
233
JAMESTOWN, ND
20
LCP
1953
1982
29
LEAKING
STRUCK WITH BACK HOE
234
JEFFERSON CO, CO
27
LCP
1976
1982
IV
LEAKING
SULFATE OR CYLINDER DEFECTS
235
JEFFERSON CO, CO
27
LCP
1976
1982
IV
RUPTURED
CORROSION
267
JUNCTION CITY, KS
20
LCP
1978
1983
IV
RUPTURED
THRUST BLOCK MOVED
236
KANSAS CITY, MO
72
ECP
1981
1982
IV
LEAKING
COATING CRACKS
405
KANSAS CITY, MO
48
LCP
1947
MB
RUPTURED
13
RUPTURED
OT
157
KNOXVILLE, TN
48
LCP
1971
1980
II
CORROSION OF PRESTRESS WIRES
(continued)
Table 1.1 continued

ID
196
Location
KNOXVILLE, TN
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
48
LCP
1972
1981
II
Failure/condition
RUPTURED
CORROSION OF WIRES; FAILURES 100 FT.

APART
KNOXVILLE, TN
48
LCP
1972
1981
II
237
LAKEHURST, CO
27
LCP
1975
1982
IV
158
LASALLE CO, ILL
54
ECP
1974
1980
IV
15
LITTLE ROCK, AR
16
LCP
1953
1971
18
RUPTURED
1
LEAKING
AGGRESSIVE SOIL
RUPTURED
DAMAGED; CHLORIDE CONTAMINATION
RUPTURED

INSTALLATION
27
197
125
LITTLE ROCK, AR
24
LCP
1965
1979
14
II
LEAKING
LOOPED GASKET
238
LITTLE ROCK, AR
48
ECP
1956
1982
26
RUPTURED
ROCK UNDER PIPE; CONTRACTOR
239
LITTLE ROCK, AR
72
ECP
1980
1982
IV
RUPTURED
240
LITTLE ROCK, AR
48
LCP
1980
1982
IV
RUPTURED
591
LIVONIA, MI
48
1972
2003
31
RUPTURED
Ruptured near Inkster Road
590
LIVONIA, MI
48
1972
2007
35
RUPTURED
Ruptured July 12, 2007
435
LOS ANGELES, CA
84
ECP
1967
II
6-
5/16
436
LOS ANGELES, CA
78
ECP
1967
II
5/16
several 10+ breaks

1
- 3/8
437
LOS ANGELES, CA
78
ECP
1967
II
1/4 -
LOS ANGELES, CA
78
ECP
1967
II
81/4
Moderately good condition; some possible 5-wire

breaks and a few 10-wire breaks
5/16
438
Moderate condition; many possible 5-wire breaks,

several 10+ breaks

several 10+ breaks
(continued)
Table 1.1 continued

ID
439
Location
LOS ANGELES, CA
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
150
ECP
1970
II
Failure/condition

breaks and a few 10-wire breaks
440
LOS ANGELES, CA
96
ECP
1970
II
1/4 -
5/16
LOS ANGELES, CA
96
ECP
1970
II
1/4

breaks and one 10-wire breaks
442
LOS ANGELES, CA
96
ECP
1970
II
6-
Good condition; one possible 5-wire break
Moderately good condition; several possible 5-
5/16
443
LOS ANGELES, CA
96
ECP
1970
II
28
441
breaks and a few 10+-wire breaks
65/16
444
LOS ANGELES, CA
84
ECP
1970
II
wire breaks
1/4 -
Moderately good condition; several possible 5-
5/16
59
LOUISVILLE, KY
60
ECP
1975
1976
IV
241
MAMARONECK, NY
30
LCP
1958
1982
24
60
MANDEN, ND
24
LCP
1975
1976
242
MANDEN, ND
30
LCP
1975
1982
243
MANDEN, ND
30
LCP
1975
244
MARTINS CREEK, PA
108
ECP
104
McGEHEE, AR
30
105
McGEHEE, AR
30
wire breaks and one 10-wire breaks

1
RUPTURED
failure at shop welded joint
BROKEN WIRES
SETTLEMENT OF CONCRETE
IV
LEAKING
LINE WAS BEING TESTED
IV
LEAKING
EXCESSIVE JOINT OPENING
1982
IV
LEAKING
CONTRACTOR; IMPROPER INSTALLATION
1973
1982
IV
RUPTURED
CONTRACTOR HIT PIPE WITH BACKHOE
LCP
1975
1978
IV
RUPTURED
SURGE DUE TO SUDDEN VALVE CLOSURE
LCP
1975
1978
IV
RUPTURED
HOLE PUNCHED BY SUPPORT BUILT ON

PIPE
80
MIAMI, FL
30
LCP
1967
1977
10
II
LEAKING
H2S
(continued)
Table 1.1 continued

ID
Location
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
23
MIDDLESEX CO., NJ
48
ECP
1969
1972
II
198
MIDDLESEX CO., NJ
102
ECP
1977
1981
III
1
8
Failure/condition
RUPTURED
CRACKED COATING
TURNBUCKLE JOINTS; ROTATION

FAILURE -BC added wire class/size per install
date (C301-79, A648)
MIDDLESEX CO., NJ
102
ECP
1977
1983
IV
335
MIDDLESEX CO., NJ
48
LCP
1967
1990
23
II
356
MIDDLESEX CO., NJ
48
LCP
1967
1992
25
II
473
MIDDLESEX COUNTY,
102
ECP
1977
1983
IV
NJ
29
268
6,
RUPTURED
COLUMN SEPARATION; SURGE
LEAKING
DENTED CYLINDER
LEAKING
CYLINDER WELD FAILURE
due to hydraulic surges exceeding the design
1/4
strength of the pipe caused by power loss to the

pumping station
472
MIDDLESEX COUNTY,
102
ECP
1977
2003
26
IV
NJ
6,
RUPTURE
1/4
the first PCCP failure in the MCUA system

attributable to environmental degradation of the
pipeline
126
MIDLAND, MI
48
LCP
1947
1979
32
37
MINOT, ND
24
LCP
1961
1974
13
RUPTURED
RUPTURED
HIGH H2S LEVELS - BC added wire size per

C301-58
106
MORRIS CO, NJ
24
LCP
1972
1978
IV
RUPTURED
SURGES CRACKED COATING
24
MORRISTOWN WTR.
16
LCP
1965
1972
II
RUPTURED
CO., NJ
INSTALLATION
81
MT LEBANON, PA
48
ECP
1958
1977
19
RUPTURED
POSSIBLE HIGH SULFATE LEVEL
159
MT LEBANON, PA
48
ECP
1958
1980
22
RUPTURED
AGGRESSIVE SOIL
520
MUSKEGON COUNTY,
66
ECP
1972
1983
11
RUPTURE
Sewer line breaks under Eastern Avenue near
MI
downtown Muskegon.
(continued)
Table 1.1 continued

ID
521
Location
MUSKEGON COUNTY,
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
66
ECP
1972
1988
16
Failure/condition
RUPTURE
Line breaks northwest of Marquette Avenue and
MI
522
MUSKEGON COUNTY,
Getty Street.
66
ECP
1972
1999
27
RUPTURE
Line breaks under Yuba Street
66
ECP
1972
2001
29
RUPTURE
Line breaks for a fourth time, under Sumner
MI
MUSKEGON COUNTY,
MI
524
MUSKEGON COUNTY,
Avenue west of Wood Street.

66
ECP
1972
2001
29
528
REPLACEMENT
MI
Muskegon County replaces a two-mile section of

66-inch sewer main, between downtown
Muskegon and Muskegon Township, near U.S. 31.
30
523
525
MUSKEGON COUNTY,
66
ECP
1972
2007
35
RUPTURE
MI
A section of 66-inch diameter sewer main,

previously thought to be structurally sound, breaks
on MacArthur Street.
107
NAGUABO, PR
30
LCP
1967
1978
11
II
RUPTURED
SURGES-7TH BREAK-IN FIRST 800' FROM

PUMP
127
NAGUABO, PR
30
LCP
1967
1979
12
II
RUPTURED
SURGE-8TH BREAK
269
NASHVILLE, TN
60
ECP
1974
1983
IV
LEAKING
POOR FIELD WELD
16
NEW BERN, NC
30
LCP
1967
1971
II
LEAKING
CORROSION
25
NEW BERN, NC
30
LCP
1967
1972
II
LEAKING
BRITTLE FRACTURE OF WIRE
26
NEW BERN, NC
30
LCP
1967
1972
II
LEAKING
COATING DAMAGED; CORE CRACKED
46
NEW BERN, NC
30
LCP
1967
1975
II
LEAKING
BRITTLE BREAKS IN WIRE
61
NEW BERN, NC
30
LCP
1967
1976
II
LEAKING
WIRES CORRODED
62
NEW BERN, NC
30
LCP
1968
1976
II
LEAKING
SEVERAL PIN HOLES & SPLIT IN CYLINDER
63
NEW BERN, NC
30
LCP
1968
1976
II
LEAKING
CORRODED WIRE; BAD DIAPER
(continued)
Table 1.1 continued

ID
Location
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
Failure/condition
NEW BERN, NC
30
LCP
1967
1978
11
II
LEAKING
128
NEW BERN, NC
30
LCP
1967
1979
12
II
LEAKING
129
NEW BERN, NC
30
LCP
1967
1979
12
II
LEAKING
130
NEW BERN, NC
30
LCP
1967
1979
12
II
LEAKING
131
NEW BERN, NC
30
LCP
1967
1979
12
II
LEAKING
1100 FT SECTION
160
NEW BERN, NC
30
LCP
1967
1980
13
II
BROKEN WIRES
COATING CRUMBLED AND WIRES

CORRODED
199
NEW BERN, NC
30
LCP
1969
1981
12
II
31
108
BROKEN WIRES;
LONGITUDINAL COATING CRACKS;
EXTERNAL
CORROSION
INSPECTION
270
NEW BERN, NC
30
LCP
1967
1983
16
II
320
NEW JERSEY WATER
108
ECP
1974
1988
14
IV
13
1
BROKEN WIRES; EXTERNAL INSPECTION

RUPTURED
SUPPLY
326
NEW JERSEY WATER
CORROSION AND BREAKAGE OF

PRESTRESSED WIRE
108
ECP
1974
1989
15
IV
SUPPLY
LINING DISTRESS,
LONGITUDINAL LINING RACKS
INTERNAL
INSPECTION
336
NEW JERSEY WATER
108
ECP
1974
1990
16
IV
SUPPLY
245
NEW YORK CITY, NY
16
LCP
1974
1982
IV
RUPTURED
BEAM PROBLEM
109
NEWARK AIRPORT,
16
LCP
1969
1978
II
RUPTURED
ELECTROLYSIS; C.P. INTERFERENCE
NJ
(continued)
Table 1.1 continued

ID
461
Location
NORTH ANDOVER,
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
72
ECP
1974
2005
31
IV
Failure/condition
14
140 pipe sections, 14 were found to have wire
MA
breaks. 4 pipe sections >50 wire breaks. In 2005,

post tensioning tendons installed on one pipe
section. Being acoustically monitored.
NORTH ANDOVER,
72
ECP
1974
IV
REPAIRED
MA
same pipeline as above. In 2005, we installed post

tensioning tendons on one
pipe section
161
NORTH GLENN, CO
48
LCP
1980
1980
IV
RUPTURED
BELL BOLT JT.; CYL. FAILED ADJ. TO

HELICAL WELD
32
445
82
NOVI, MI
24
LCP
1975
1977
IV
459
OAKLAND, CA
162
OKLAHOMA CITY, OK
60
ECP
1973
1980
IV
200
OKLAHOMA CITY, OK
60
ECP
1973
1981
IV
246
OKLAHOMA CITY, OK
60
ECP
1973
1982
IV
271
OKLAHOMA CITY, OK
60
ECP
1973
1983
10
272
OKLAHOMA CITY, OK
60
ECP
1973
1983
273
OKLAHOMA CITY, OK
60
ECP
1973
290
OKLAHOMA CITY, OK
60
ECP
291
OKLAHOMA CITY, OK
60
308
OKLAHOMA CITY, OK
60
LEAKING
LOOPED GASKET
4/7/86/ PBC list
BROKEN WIRES
LOOPED GASKET OR JOINT LEAK
RUPTURED
SURGED
RUPTURED
SURGES
IV
RUPTURED
SURGED
10
IV
BROKEN WIRES
SURGED
1983
10
IV
RUPTURED
SURGED
1973
1984
11
IV
RUPTURED
ECP
1973
1984
IV
RUPTURED
ECP
1974
1987
13
IV
RUPTURED
STEEL CYLINDER RUPTURED
(continued)
Table 1.1 continued

ID
309
Location
OKLAHOMA CITY, OK
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
60
ECP
1974
1987
13
IV
Failure/condition
LINING DISTRESS,
LEAKING CYLINDER
INTERNAL
INSPECTION
OKLAHOMA CITY, OK
60
ECP
528
OKLAHOMA CITY, OK
72
ECP
83
OMAHA, NE
36
LCP
1974
2004
30
IV
2006
1974
1977
IV
RUPTURED
RUPTURES
RUPTURED
TORN ALONG HELICAL; BELL BOLT JTS. 45

ELBOW
110
OMAHA, NE
36
LCP
1974
1978
IV
RUPTURED
TORN ALONG HELICAL; BELL BOLT JTS. 45

ELBOW
33
563
163
OMAHA, NE
48
LCP
1974
1980
IV
RUPTURED
SHORT BETWEEN TWO ELBOWS W/BELL

BOLT JT.
164
OMAHA, NE
48
LCP
1974
1980
ONONDAGA CO., NY
30
LCP
1947
1960
23
IV
1
1/8

CRACKED
CHEMICAL WASTES
COATING
84
ONONDAGA, NY
30
LCP
1976
1977
IV
RUPTURED
SEAM RUPTURE DURING TESTING; PIPE ON

ROCK
327
ORADELL, NJ
554
ORLANDO, FL
60
ECP
36, 42
1988
1989
III
1984
2004
20
III
LEAKING
WELDED RESTRAINED JOINT PARTED
RUPTURE
Failed several times. Replaced with BA04-89,

BR04-289, Project No. 2729 -BC added wire class
& size per year installed (C301-79 or 84, A648)
132
PALM BEACH, FL
20
LCP
1956
1979
23
BROKEN WIRES
H2S; H2S04
(continued)
Table 1.1 continued

ID
476
Location
PHOENIX, AZ
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
252
ECP
1978
1990
12
III
1/2
91
Failure/condition
REPAIR
corrosion monitoring indicated several pipes along

all six siphons corroding. OF the 223 pipe units
excavated to springline, 91 (41%) exhibited
external distress which required additional repairs
PHOENIX, AZ
252
ECP
1978
1990
12
III
1/2
35
REPAIR
Relining the first 700 ft of siphon under Interstate

10 with steel
478
PHOENIX, AZ
252
ECP
1978
1990
12
III
1/2
25
REPAIR
Relining 500 ft of siphon with steel
480
PHOENIX, AZ
252
ECP
1978
1993
15
III
1/2
425
REPLACEMENT
replacing the existing siphon with parallel cast-inplace concrete or steel pipelines (8,492 linear feet)
34
477
481
PHOENIX, AZ
252
ECP
1978
1993
15
III
1/2
278
REPLACEMENT
363
PHOENIX, AZ
66
ECP
1978
1994
16
II
RUPTURED
389
PHOENIX, AZ
66
ECP
1978
1997
19
II
RUPTURED
390
PHOENIX, AZ
96
ECP
1978
1997
19
II
RUPTURED
485
PHOENIX, AZ
72-
ECP
1973
2003
30
3,696
The City of Phoenix initiated a multi-year
108
rehabilitation program to slipline all but one mile

of the pipeline using steel pipe.
452
PHOENIX, AZ
60
ECP
1978
479
PHOENIX, AZ
252
ECP
1978
2006
28
III
III
1/2
Failure in October 2006 (the failure was in the section as-built in 1978)
492
REPLACEMENT
133
PINELLAS CO ,FL
60
ECP
1977
1979
IV
RUPTURED
COATING AND CORE DAMAGED PBC list

indicates 1978 construction
(continued)
Table 1.1 continued

ID
165
Location
PINELLAS CO ,FL
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
60
ECP
1977
1980
IV
Failure/condition
RUPTURED
BEING REFILLED FOR STATIC TESTING;

SURGE PBC list indicates 1978 construction
406
PINELLAS CO ,FL
60
482
PINELLAS COUNTY,
48
ECP
1977
IV
RUPTURED
2001
No rupture
Accoustically monitored by PURE, found several

problematic areas. Jason Consultants
recommended sliplining.
35
FL
85
PITTSBURGH, PA
48
LCP
1977
RUPTURED
COATING DETERIORATION
166
PITTSBURGH, PA
48
LCP
1980
RUPTURED
COATING DETERIORATION; SULFATES
274
PITTSBURGH, PA
48
LCP
1983
292
PITTSBURGH, PA
36
LCP
1960
1984
24
RUPTURED
AT JOINT
300
PITTSBURGH, PA
36
LCP
1960
1985
25
RUPTURED
MID-PIPE AT BOTTOM
275
PLEASANT PRAIRE,
30
LCP
1978
1983
IV
RUPTURED
30
LCP
1978
1983/
IV
RUPTURED
26
III
LEAKING
WI
402
PLEASANT PRAIRE,
WI
377
PROVIDENCE, RI
2ND PROBLEM; 5 FAILURES BY 1985
5
102
LCP
1970
1996
RUPTURED
ALSO ADJACENT PIPE REPLACED -BC added

wire class & size per installed year (AWWA C30164, A227)
557
PROVIDENCE, RI
102
LCP
1970
2000
30
III
11
REPAIRED
-BC added wire class & size per installed year

(AWWA C301-64, A227)
86
PUERTO RICO
60
ECP
276
RALEIGH, NC
36
LCP
1977
1974
1983
1
9
IV
RUPTURED
1
LEAKING
INSTALLED ON POOR FOUNDATION
(continued)
Table 1.1 continued

ID
201
Location
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
1970
1981
11
III
RIALTO, CA
Failure/condition
RUPTURED
-BC added wire class/size per install date (C30164, A227)
429
RIALTO, CA
96
ECP
1972
2004
32
III
8-
REPAIRED
Moderate condition; several break locations
5/16
RIALTO, CA
120
ECP
1972
III
6-
Moderate condition; several 5-wire break locations
5/16
36
430
277
RICHMOND, VA
36
LCP
1982
1983
IV
310
RICHMOND, VA
36
LCP
1979
1987
IV
321
RICHMOND, VA
36
LCP
1979
1988
IV
426
RIVERSIDE, CA
96
ECP
1976
III
6-8
27
ROCHESTER, NY
20
LCP
1960
1972
12
134
ROCK HILL, SC
30
LCP
1972
1979
IV
87
S.E. OAKLAND CO., MI
20
LCP
1970
1977
II
LOOSE COATING
10 PIPE REPAIRED
RUPTURED
WIRE NOT AS SPECIFIED
RUPTURED
WIRE NOT AS SPECIFIED
1
1
No broken wires; however, several "broken backs"

RUPTURED
SFM DAMAGED DURING INSTALLATION
LEAKING
TEAR IN WELD AT JT. RING; PIPE REPAIRED
LEAKING
STORM SEWER FLUSHING LINE W/4 12"

OUTLETS
88
20
LCP
1975
1977
IV
LEAKING
STORM SEWER FLUSHING LINE W/4 12"

OUTLETS
204
20
LCP
1970
1981
11
II
460
SACRAMENTO, CA
135
SAGINAW, MI
48
LCP
1948
1979
31
293
SALT LAKE CITY, UT
66
ECP
1981
1984
III
LEAKING
4/7/86 PBC list indicates 1 failure
1
8
SPLIT WELD
RUPTURED
1
RUPTURED
BROKEN WIRES, CATHODIC PROTECTION Two units of pipe (failed No. 797 and neighbor
798) were replaced
(continued)
Table 1.1 continued

ID
Location
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
1981
1990
III
553
SALT LAKE CITY, UT
66
ECP
SAN DIEGO CO. CA
24
LCP
136
SAN DIEGO CO. CA
66-
ECP
1962
1960
1979
19
607
Failure/condition
RELINED
All 2.3 miles were rehabilitated by lining with steel
RUPTURED
RUPTURED
69
1979
SAN DIEGO CO. CA
72
ECP
1972
1979
III
567
SAN DIEGO CO. CA
66,
ECP
1960
1980
20
RUPTURED
Broken back at structure (date approximate)
RELINED
Rupture Dec 31, 1980. BC added wire class/size
69
SAN DIEGO CO. CA
66
per install year (C301-58, A227)

ECP
1960
1982
22
1,322
RELINED
Relined 1982-84. BC added wire class/size per

install year (C301-58, A227)
37
248
462
FAILURE OF 66-69 INCH PIPE. Rupture Feb 23,
568
SAN DIEGO CO. CA
66,
ECP
1960
1982
22
RELINED
69
Rupture May 28, 1982. BC added wire class/size

per install year (C301-58, A227)
251
SAN DIEGO CO. CA
72
ECP
1972
1982
10
III
RUPTURED
9/18/82 BROKEN WIRES; CORROSION
278
SAN DIEGO CO. CA
72
ECP
1973
1983
10
III
RUPTURED
BROKEN WIRES/ CORROSION
249
SAN DIEGO CO. CA
72
ECP
1972
1987
15
III
RUPTURED
8/3/87 replaced 8 sections
247
SAN DIEGO CO. CA
72
ECP
1959
1988
29
III
RUPTURED
January - 1 pipe replaced
250
SAN DIEGO CO. CA
72
ECP
1972
1990
18
III
RUPTURED
replaced 1 section
337
SAN DIEGO CO. CA
84
ECP
1974
1990
16
III
RUPTURE,
Rupture October 1990. External inspection found
EXTERNAL
severe corrosion on pipe 547
INSPECTION
569
SAN DIEGO CO. CA
84
ECP
1974
1991
17
III
LINING DISTRESS,
INTERNAL
INSPECTION
(continued)
Table 1.1 continued

ID
570
Location
SAN DIEGO CO. CA
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
96
ECP
1972
1992
20
Failure/condition
LINING DISTRESS,
INTERNAL
INSPECTION
571
SAN DIEGO CO. CA
66
ECP
1972
1992
20
LINING DISTRESS,
INSTALLATION
572
SAN DIEGO CO. CA
96
ECP
1972
1993
21
RUPTURE,
BROKEN WIRES/ CORROSION. Rupture Nov.
EXTERNAL
1993
INSPECTION
38
P4EI
573
574
575
SAN DIEGO CO. CA
SAN DIEGO CO. CA
SAN DIEGO CO. CA
66
69
72
ECP
ECP
ECP
1960
1959
1977
1994
1997
1997
34
38
20
INTERNAL
BROKEN WIRES/ CORROSION. BC added wire
INSPECTION
class/size per install year (C301-58, A227)
INTERNAL
MANUFACTURING DEFECT. BC added wire
INSPECTION
INTERNAL
Internal corrosion, cracked can, leaking water
INSPECTION
576
577
SAN DIEGO CO. CA
SAN DIEGO CO. CA
66
69
ECP
ECP
1964
1959
1999
1999
35
40
II
CONCRETE
BC added wire class/size per install year (C301-64,
ENCASED
A227)
INTERNAL
BROKEN WIRES/ CORROSION. Relined. BC
INSPECTION
added wire class/size per install year (C301-58,

A227)
578
SAN DIEGO CO. CA
84
ECP
1974
2003
29
III
LINING DISTRESS,
BROKEN WIRES/ CORROSION from RFEC/TC
INTERNAL
INSPECTION
584
SAN DIEGO CO. CA
69
ECP
1959
2003
44
851
RELINED
Relined. BC added wire class/size per install year

(C301-58, A227)
(continued)
Table 1.1 continued

ID
579
Location
SAN DIEGO CO. CA
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
96
ECP
1972
2004
32
Failure/condition
INTERNAL
BROKEN WIRES/ CORROSION from RFEC/TC
INSPECTION
585
SAN DIEGO CO. CA
96
ECP
1972
2004
32
580
SAN DIEGO CO. CA
69
ECP
1959
2004
45
RELINED
Relined 2003-2004
11
INTERNAL
BROKEN WIRES/ CORROSION. PPIC
INSPECTION
RFEC/TC. Concrete encased. BC added wire

39
835
586
SAN DIEGO CO. CA
96
ECP
1972
2004
32
1,325
RELINED
Relined 2004-2005
587
SAN DIEGO CO. CA
96
ECP
1972
2004
32
1,071
RELINED
Relined 2004-2005
581
SAN DIEGO CO. CA
90
ECP
1969
2005
36
INTERNAL
BROKEN WIRES/ CORROSION. BC added wire
INSPECTION
INTERNAL
LONGITUDINAL COATING CRACKS;
INSPECTION
CORROSION
RELINED
Relined 2005-2006
RUPTURE
Failure 5-16-06 on Mission Trails -BC added wire
582
SAN DIEGO CO. CA
33
LCP
1978
2005
III
27
588
SAN DIEGO CO. CA
96
ECP
1972
2005
33
463
SAN DIEGO CO. CA
66
ECP
1959
2006
47
266
I

465
SAN DIEGO CO. CA
96
ECP
1981
2006
25
III
REPAIRED
Pieces 234 & 240 repaired with Carbon Fiber June

2006 after eddy current testing -BC added wire
class/size per install year (C301-79, A648
464
SAN DIEGO CO. CA
96
ECP
1981
2007
26
III
REPAIRED
Pieces 1225 & 2196 repaired with Carbon Fiber

March 2007 after eddy current testing -BC added
wire class/size per install year (C301-79, A648
583
SAN DIEGO CO. CA
33,
42
LCP
1978
2007
29
INTERNAL
Leak detected after moisture surfaced.
INSPECTION
(continued)
Table 1.1 continued

ID
425
Location
SAN DIEGO CO. CA
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
96
ECP
1985
III
Failure/condition
REPAIRED
Pipe includes longitudinal shorting strap under

wires. Only one break area caused by mechanical
damage, otherwise in good condition
432
SAN DIEGO CO. CA
99
ECP
1970
II
8-
Moderate good condition; a few 5-wire breaks
433
SAN DIEGO CO. CA
96
ECP
1981
III
8-
Good condition; one 10-wire break
5/16
338
SAN FRANCISCO, CA
66
ECP
1980
1990
10
III
40
5/16
347
SAN FRANCISCO, CA
84
ECP
1985
1991
II
LINING DISTRESS,
Failure @ 153+40. Investigated by CERCO
INTERNAL
(No.9015). Hydrogen Embrittlement suggested as
INSPECTION
cause.
LINING DISTRESS,
CARBONATED COATING
INTERNAL
INSPECTION
532
SAN FRANCISCO, CA
96
ECP
1966
2004
38
II
REPAIR
Repair four brittle fractures in pipe 749 after PPIC

inspection -BC added wire class/size per install
year (C301-64, A227)
434
SAN JACINTO, CA
96
ECP
1975
III
Only one break area caused by mechanical

damage, otherwise in good condition
47
SANFORD, NC
24
LCP
1971
1975
II
202
SANFORD, NC
24
LCP
1971
1981
10
II
203
SANFORD, NC
24
LCP
1981
1981
IV
446
SANTA CLARA, CA
96
ECP
1985
III
6,
1/4,
LEAKING
DAMAGED
RUPTURED
LARGE BOULDER; CONTRACTOR DAMAGE
RUPTURED
ELECTROLYSIS
About 10% inspected (eddy current) and found to be in excellent condition

(Fall 2003)
5/16
(continued)
Table 1.1 continued

ID
279
Location
SAYERVILLE,NJ
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
102
ECP
1980
1983
IV
Failure/condition
BROKEN WIRES
POSSIBLE WIRE BRIDGING AND FAULTY

INSTALLATION
31
SOUTH
24
LCP
1952
1973
21
48
ECP
1958
1983
25
RUPTURED
WRONG CLASS PIPE USED; HIGH PRESSURE
LEAKING
PIPE SETTLED
PITTSBURGH,PA
SOUTH
PITTSBURGH,PA
41
280
17
ST. LOUIS CO., MO
36
LCP
1952
1971
19
LEAKING
MANUFACTURING DEFICIENCIES
18
ST. LOUIS CO., MO
36
LCP
1952
1971
19
LEAKING
MANUFACTURING DEFICIENCIES
38
ST. LOUIS CO., MO
LCP
1950
1974
24
RUPTURED
MANUFACTURE DEFECT
39
ST. LOUIS CO., MO
36
LCP
1952
1974
22
1 LEAKING; 1 WITH BROKEN WIRES
64
ST. LOUIS CO., MO
36
LCP
1952
1976
24
1 LEAKING; 1 WITH BROKEN WIRES
137
ST. LOUIS CO., MO
36
LCP
1953
1979
26
550
ST. PAUL, MN
48
1972-78
IV
1
8
RUPTURED
REPLACEMENT
exhibited cracks in the protective cement mortar

coating, delaminated protective cement mortar
coating, and corroded and broken prestressed
wires. -BC added install date & wire size per wire
class/manufacturer (C301-72, A648, Interpace
practice)
281
STAMFORD, CT
24
LCP
SYRACUSE, NY
30
LCP
1969
1983
1961
14
II
I
RUPTURED
RUPTURED
STRAY CURRENT; SEVERAL FAILURES
(continued)
Table 1.1 continued

ID
498
Location
TAMPA, FL
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
66
ECP
1975
1987
12
IV
Failure/condition
RUPTURE
A pipeline investigation was conducted between

February 1990 and April 1992 by PURE. Soils
and groundwater chemistry were generally benign.
-BC added wire class and size per install date
(C301-72, A648, Interpace practice)
TAMPA, FL
84
ECP
1975
1990
15
IV
RUPTURE

(C301-72, A648, Interpace practice)
357
TAMPA, FL
90
ECP
1975
1992
17
IV
500
TAMPA, FL
84
ECP
1975
1997
22
III
1
1
EXTERNAL INSPECTION; BROKEN WIRES

RUPTURE
Rupture October 30, 1997 in an area intermixed

with Interpace pipe -BC added wire class and size
42
499
per install date (C301-72, A648)

561
TAMPA, FL
84
ECP
1975
1998
23
IV
RUPTURE
Rupture September 23, 1998 -BC added wire class

and size per install date (C301-72, A648, Interpace
practice)
562
TAMPA, FL
84
ECP
1975
1999
24
IV
RUPTURE
Rupture April 22, 1999 -BC added wire class and

size per install date (C301-72, A648, Interpace
practice)
502
TAMPA, FL
84
ECP
1975
1999
24
III
Conclusion from acoustic monitoring: Sta. 673+55

& 683+35 were repaired/replaced. -BC added
wire class and size per install date (C301-72,
A648)
501
TAMPA, FL
84
ECP
1975
2000
25
IV
2,000
REPLACEMENT
Conclusion from internal inspection: the Interpace

piping should be immediately replaced. 40,000
feet replaced with Price Brothers PCCP in fall of
2000 -BC added wire class and size per install
date (C301-72, A648, Interpace practice)
(continued)
Table 1.1 continued

ID
559
Location
TAMPA, FL
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
84
ECP
1975
2003
28
III
Failure/condition
RUPTURE
Second rupture of Price Brother's pipe on March

20, 2003. Three additional pieces removed. -BC
added wire class and size per install date (C30172, A648)
TAMPA, FL
84
ECP
1975
2003
28
III
1,628
REPLACEMENT
After second rupture, Board boted to replace the

remaining 5.5 mile segment -BC added wire class
and size per install date (C301-72, A648)
555
TAMPA, FL
84
ECP
1975
2005
III
1,450
PARALLEL MAIN
The new main parallels and will replace the final

section of a failing 84 main. -BC added wire
class & size per install date (C301-72, A648)
43
560
504
505
506
511
507
508
509
TARRANT COUNTY,
72,
TX
84
TARRANT COUNTY,
72,
TX
84
TARRANT COUNTY,
72,
TX
84
TARRANT COUNTY,
90,
TX
108
TARRANT COUNTY,
72,
TX
84
TARRANT COUNTY,
72,
TX
84
TARRANT COUNTY,
72,
TX
84
ECP
1972
1981
II
6, 8,
RUPTURE
Corrosion failure
RUPTURE,
Corrosion failure
1/4
ECP
1972
1983
11
II
6, 8,
1/4
ECP
1972
1988
16
II
6, 8,
REPAIRED
1
RUPTURE,
1/4
ECP
1988
1988
III
REPAIRED
6,
Thrust failures during hydrostatic testing and first
1/4
ECP
1972
1989
17
II
6, 8,
pump test
1
1/4
ECP
1972
1990
18
II
6, 8,
1972
1991
19
II
6, 8,
1/4
RUPTURE,
Corrosion failure
REPAIRED
1
1/4
ECP
Corrosion failure
RUPTURE,
Corrosion failure
REPAIRED
2
RUPTURE,
Corrosion failures
REPAIRED
(continued)
Table 1.1 continued

ID
512
513
510
515
44
514
516
517
518
592
Location
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
Failure/condition
(in.)
type
date
date
yrs.
class
size
ruptures
failures
TARRANT COUNTY,
90,
ECP
1988
1992
III
6,
RUPTURE
Corrosion failure
TX
108
TARRANT COUNTY,
90,
RUPTURE
Thrust restraint failure
TX
108
TARRANT COUNTY,
90,
RUPTURE
Thrust restraint failure
TX
108
TARRANT COUNTY,
72,
RUPTURE,
Hydrogen embrittlement failure
TX
84
TARRANT COUNTY,
90,
TX
108
TARRANT COUNTY,
90,
TX
108
TARRANT COUNTY,
90,
TX
108
TARRANT COUNTY,
90,
TX
108
TARRANT COUNTY,
72
1/4
ECP
1988
1993
III
6,
1/4
ECP
1988
1995
III
6,
1/4
ECP
1972
1996
24
II
6, 8,
1/4
ECP
1988
1996
III
6,
REPAIRED
2
RUPTURE
1/4
ECP
1988
1997
III
6,
failure
1
RUPTURE
Corrosion failure
RUPTURE
Corrosion failure and hydrogen embrittlement
1/4
ECP
1988
1998
10
III
6,
1/4
ECP
1988
1999
11
III
6,
failure
1
RUPTURE
Corrosion failure
RUPTURE,
Failed January 2004
1/4
ECP
1972
2004
32
II
TX
519
431
REPAIRED
TARRANT COUNTY,
72,
TX
84,
TEMECULA, CA
Corrosion failure and hydrogen embrittlement
ECP
1972
2005
33
III
6,
70
1/4
REPAIR/REPLACEM
Results of the PPIC inspections revealed that of the
ENT
37148 segments inspected, 818 segments have
90,
wire breaks. Repairs were made choosing pipes
108
with 50 or more broken wires.
162
ECP
1971
II
1/4 -
Moderate condition; several break locations
5/16
(continued)
Table 1.1 continued

ID
Location
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
45
252
TROY, MI
16
LCP
1974
1982
IV
398
TUCSON, AZ
96
ECP
1988
1999
11
III
483
TUCSON, AZ
96
ECP
205
URBANA, IL
30
LCP
1964
1981
17
89
UTE, CO
18
LCP
1964
1977
111
UTE, CO
18
LCP
1964
138
UTE, CO
24
NCP
139
UTE, CO
24
UTICA, NY
253
1
6
Failure/condition
LEAKING
ROLLED GASKET; CONTRACTOR
RUPTURED
RUPTURE, corrosive soil
II
RUPTURED
13
II
1978
14
II
LEAKING
1964
1979
15
II
LEAKING
LOOPED GASKET
NCP
1964
1979
15
II
RUPTURED
MORTAR COATING DETERIORATION
30
LCP
1954
1967
13
RUPTURED
OPEN JOINT OR LOOPED GASKET
WALWORTH CO, WI
20
LCP
1981
1982
IV
LEAKING
CONTRACTOR LEFT 1" JOINT SEPARATION
10
WASHINGTON DC
30
LCP
1962
1969
LEAKING
254
WAYNE, NJ
1981
1982
IV
RUPTURED
CONTRACTOR; COLLAPSED CORE
311
WEST COAST
RUPTURED
CORRODED WIRES; POROUS COATING
BROKEN WIRES
OTL SURVEY
RUPTURED
WIRE CORROSION
RUPTURED
1999
84
ECP
1974
1987
13
IV
84
ECP
1974
1991
17
IV
84
ECP
1974
1991
17
IV
27
LCP
1965
1980
15
II
CLASS 90 SURCHARGE LOAD
LEAKING
REGIONAL, FL
348
WEST COAST
REGIONAL, FL
349
WEST COAST
REGIONAL, FL
167
WEST ORANGE, NJ
(continued)
Table 1.1 continued

ID
503
Location
WESTFIELD, NJ
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
60
ECP
1977
2002
25
IV
Failure/condition
RUPTURE
Investigation of the leak determined that the cause

was corrosion of the steel cylinder where the
concrete lining had been cut back to weld the joint
and the lining had not been repaired. External
inspection found one additional prob. pipe
46
255
WESTMORELAND, PA
20
LCP
1956
1982
26
LEAKING
ACID ATTACK
282
WHEATLAND, WY
24
LCP
1978
1983
IV
RUPTURED
GRADE E SHEET; EXCESS LOAD
140
WILMINGTON, DE
24
LCP
1970
1979
II
RUPTURED
H2S; H2SO4
141
WILMINGTON, NC
24
LCP
1970
1979
II
RUPTURED
H2S; H2SO4
32
WINSTON-SALEM, NC
36
LCP
1949
1973
24
RUPTURED
SURGES
486
WINTERSBERG, AZ
114,
ECP
1982
1994
12
III
RUPTURE
96,

(C301-79, A648)
66
487
WINTERSBERG, AZ
114,
ECP
1982
1995
13
III
Accoustically monitored by PURE, found five pipe
96,
sections with acoustic events. Two of the pipe
66
sections had indications of intermediate to

advanced distress and were repaired. -BC added
wire class and size per install date (C301-79,
A648)
11
WSSC, MD
16
LCP
1965
1970
II
3 LEAKING; 3
W/BROKEN WIRES
48
WSSC, MD
30
LCP
1943
1975
32
49
WSSC, MD
96
ECP
1970
1975
II
1
6
LEAKING
ERODED CORE
RUPTURED
CORRODED WIRE; PRIOR DAMAGE TO

COATING
(continued)
Table 1.1 continued

ID
112
Location
WSSC, MD
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
84
ECP
1974
1978
IV
Failure/condition
RUPTURED
DAMAGED COATING AND CORRODED

WIRES
142
WSSC, MD
42
LCP
1974
1979
IV
RUPTURED
BROKEN CYL.; DAMAGED AT

MANUFACTURE
47
168
WSSC, MD
60
ECP
1967
1980
13
II
RUPTURED
BROKEN WIRE; POSSIBLE SURGE
169
WSSC, MD
60
ECP
1967
1980
13
II
RUPTURED
BROKEN WIRE; POSSIBLE SURGE
170
WSSC, MD
72
ECP
1974
1980
IV
RUPTURED
COATING DAMAGED ON ONE SIDE
171
WSSC, MD
84
ECP
1974
1980
IV
RUPTURED
172
WSSC, MD
72
ECP
1976
1980
RUPTURED
CYLINDER AND PRESTRESS WIRES

CORRODED
206
WSSC, MD
42
LCP
1971
1981
10
II
LEAKING; BROKEN
MANUFACTURER-DENTED CYLINDER;
WIRES
SPLIT
RUPTURED
BELL JOINT RING OUT-OF-ROUND
283
WSSC, MD
42
LCP
1974
1983
IV
284
WSSC, MD
42
LCP
1974
1983
IV
294
WSSC, MD
60
ECP
1967
1984
II
RUPTURED
DISRUPTED COATING
295
WSSC, MD
84
ECP
1974
1984
10
IV
RUPTURED
LEAKING CYLINDER
296
WSSC, MD
72
ECP
1974
1984
10
IV
BROKEN WIRES,
LEAKING CYLINDER
2 WITH BROKEN WIRES; 1 ROLLED GASKET
EXTENSIVE
CORROSION
304
WSSC, MD
60
ECP
1967
1986
19
II
RUPTURED
LEAKING JOINT RING WELD
305
WSSC, MD
84
ECP
1975
1986
11
IV
1/4
RUPTURED
CORROSION; WIRE BREAKAGE
(continued)
Table 1.1 continued

ID
306
Location
WSSC, MD
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
42
LCP
1975
1986
11
IV
Failure/condition
RUPTURED
WIRE BREAKAGE DUE TO DENTED

CYLINDER
312
WSSC, MD
60
ECP
1967
1987
20
II
RUPTURED
CORROSION; BROKEN WIRE; WIRE

EXPOSED
WSSC, MD
48
ECP
1970
1987
17
II
RUPTURED
LEAKING THROUGH WELDING

IMPERFECTIONS
314
WSSC, MD
42
LCP
1975
1987
12
IV
RUPTURED
SPIGOT END LEAKING RESULTED IN

BURST
48
313
315
WSSC, MD
42
LCP
1975
1987
12
IV
RUPTURED
DENTED CYLINDER
316
WSSC, MD
30
LCP
1975
1987
12
IV
RUPTURED
CYLINDER SPLIT AT SPIGOT END
322
WSSC, MD
24
LCP
1966
1988
22
II
RUPTURED
DISRUPTED MORTAR COATING
328
WSSC, MD
96
ECP
1966
1989
23
II
329
WSSC, MD
60
ECP
1967
1989
22
II
LINING DISTRESS,
INTERNAL LONGITUDINAL CRACKS;
INTERNAL
BROKEN WIRES
INSPECTION
339
WSSC, MD
42
LCP
1959
1990
31
340
WSSC, MD
48
LCP
1960
1990
30
II
1 LEAKING; 3 W/BROKEN WIRES
EXTERNAL
BROKEN WIRES
INSPECTION
341
WSSC, MD
24
LCP
1972
1990
18
IV
RUPTURED
358
WSSC, MD
42
LCP
1974
1992
18
IV
RUPTURED
364
WSSC, MD
42
LCP
1974
1994
20
IV
RUPTURED
(continued)
Table 1.1 continued

ID
365
Location
WSSC, MD
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
108
ECP
1977
1994
18
IV
Failure/condition
LINING DISTRESS,
BROKEN WIRES
INTERNAL
INSPECTION
49
371
WSSC, MD
60
ECP
1960
1995
35
II
RUPTURED
AT SPIGOT HAUNCH; HALF BEVEL
372
WSSC, MD
60
ECP
1967
1995
28
II
RUPTURED
CYLINDER SPLIT AT BEVEL SADDLE WELD
373
WSSC, MD
36
LCP
1967
1995
28
II
RUPTURED
374
WSSC, MD
108
ECP
1977
1995
18
IV
378
WSSC, MD
60
ECP
1967
1996
29
II
LINING DISTRESS,
INTERNAL LONGITUDINAL CRACKS;
INTERNAL
BROKEN WIRES
INSPECTION
379
WSSC, MD
96
ECP
1970
1996
26
II
380
WSSC, MD
96
ECP
1970
1996
26
II
1/4
1
1
RUPTURED
LEAKING CYLINDER
LINING DISTRESS,
CORROSION; WIRE BREAKAGE
INTERNAL
INSPECTION
381
WSSC, MD
24
LCP
1966
1996
30
II
RUPTURED
DISRUPTED COATING; AUGER DAMAGED
399
WSSC, MD
54
ECP
1967
1999
32
II
400
WSSC, MD
54
ECP
1965
1999
34
II
401
WSSC, MD
54
ECP
1965
1999
34
II
(continued)
Table 1.1 continued

ID
589
Location
YORBA LINDA, CA
Dia.
Pipe
Install
Fail
Age
Wire
Wire
No. of
Other
(in.)
type
date
date
yrs.
class
size
ruptures
failures
69
ECP
1979
1999
20
III
1/4
Failure/condition
RUPTURED
Ruptured December 13, 1999 near station 187+80.

Brittle wire.
LCP = LINED CYLINDER PIPE
NCP = NON-CYLINDER
MBOT = MILD
RFEC/TC*: Remote Eddy Current/Transformer
PIPE
BRIGHT OIL
Coupling
592
TOTAL
ECP = EMBEDDED CYLINDER PIPE
DW = DOUBLE
FAC = Fly Ash Cement
WRAPPED
50
TEMPERED WIRE
(continued)
51
Figure 1.1 Location of failures LC-PCCP
52
Figure 1.2 Location of failures EC-PCCP
CHAPTER 2
HISTORY OF PCCP
PRODUCTION DATA
Early reinforced concrete pressure pipe was an extension of concrete culvert pipe, with
addition of a steel cylinder for water retention. Spirally wound reinforcement has been reported
as early as 1929 (Lynch, Wilson, and Gunther 2004) for concrete pressure pipe. In response to
the high demand for steel in the war years, the emerging technology of prestressing concrete was
applied to water pipelines. The earliest application in the U.S. was in 1942 (AWWA C301-99
Foreword). That pipe, which is now known as LC-PCCP consists of a steel cylinder with cast
concrete core, over-wrapped with steel wire. Continuing scarcity of steel, coupled with
successful application to smaller pipe, led to the introduction of EC-PCCP in 1953 (AWWA
C301-99 Foreword). Site-manufactured EC-PCCP has been constructed as large as 256 inches in
diameter.
Significant differences in the manufacture of LC-PCCP and EC-PCCP not only exist, but
the configuration of each type has changed significantly throughout the years of manufacture.
Table 2.1 lists some of the major differences in the pipe, and Table 2.2 lists some of the
significant changes in the products. Those changes are described in more detail below.
Table 2.1
Differences between lined and embedded type PCCP
Feature
Diameter range
Construction
Prestressing wire
LC-PCCP
16 through 60 inches
Steel cylinder lined with a cast concrete core
Wrapped over steel cylinder
EC-PCCP
30 through 256 inches
Steel cylinder embedded in a concrete core
Wrapped over concrete core
Table 2.2
PCCP significant changes with time
Date/Event
1942
1949
1952
1953
1955
1958
1964
1972
1979
1979
1984
Revision
First installation of LC-PCCP in U.S.
First edition of AWWA C301-Tentative, allowable wire stress approximately 45% of
ultimate strength and minimum mortar coating thickness 7/8 inch.
First edition of AWWA C301.
First installation of EC-PCCP in U.S.
Tentative standard. Included minimum design basis.
Second edition of AWWA C301, allowable wire stress 70% of ultimate strength and
minimum mortar coating thickness 5/8 inch.
Third edition of AWWA C301, combined loading design procedure added; allowable wire
stress 75% of ultimate strength.
AWWA C301 revised.
AWWA C301 revised.
Manual M9, first edition.
AWWA C301 revised, minimum mortar coating increased to 3/4 inch; cast concrete coating
deleted.
(continued)
53
Table 2.2 (continued)

Date/Event
1992
1995
1999
2007
2007
Revision
AWWA C301 revised, design appendices deleted, minimum wire size increased to 0.192
inch, minimum cylinder thickness increased to 16 gauge. First edition of AWWA C304.
Manual M9, second edition.
AWWA C301 revised.
AWWA C301 revised.
Manual M9, third edition.
PCCP was first produced and installed in the U.S. in 1942. Since that time, more than
100,000,000 feet of pipe have been produced. In order to characterize the population of PCCP
production on an annual basis and allow evaluation of the effects of changes in production,
records were solicited. Production data for Ameron Concrete Pipe Group were furnished by
Henry Bandalgian. Production data for Cretex & Price Brothers were furnished by Armand
Tremblay. Mr. Tremblay also provided production data from Lockjoint and its successors
Interpace and GHA-Lockjoint. Production data from Hanson Pipe & Precast were furnished by
Sam Arnaout. Production data for United Concrete Pipe were determined from records furnished
by Bob Card and Jim Davenport. Those data are listed in Table 2.3 and illustrated in Figures 2.1
and 2.2.
Production data for Hyprescon, Inc., and Vianini were not made available to the project.
Hydro Conduit Corp. constructed a plant in Corona, California in 1968 to manufacture EC-PCCP
for western water utilities. According to Joe Zaccarro, Hydro Conduit chief engineer, the
manufacturing records have been lost due to two relocations of the engineering department.
Hydro Conduits production ended in the mid-1970s when the equipment was sold. Some PCCP
made by LaFarge was also installed in the U.S.
Total production by year of both ECP and LCP is compared in Figure 2.3. Production of
EC-PCCP began in 1951. The low in production occurred the following year, with only 950
linear feet manufactured. The peak year production was 1984, with 1,936,970 linear feet
manufactured. Total production of ECP by manufacturer is illustrated in Figure 2.4. Production
of LC-PCCP continues today. From a low of 21,000 linear feet of pipe produced in 1942 to a
high of 2,433,523 feet in 1971, there has been at least 71,638,359 feet produced, or about 72
percent of all PCCP production. Total production of LCP by manufacturer is illustrated in
Figure 2.5. Peak production for both types of PCCP occurred in 1971 with 3,563,327 feet
produced.
Please note that because of the differences in reporting and recording production data,
pipe could have been booked as a sale one year, manufactured in the following year, and
installed the next year. Only analysis of more complete data from the utility could refine the
numbers and is unlikely to result in statistically different information.
54
CHANGES IN PIPE STANDARDS

The first PCCP was manufactured by LockJoint in 1942 to its own design for the Naval
Oil Depot in Virginia. Production continued under the manufacturers standard until 1949 when
the first AWWA Consensus Standard 7B.2-T was tentatively adopted on November 21, 1949.
That was followed in 1952 by adoption of the first edition of AWWA C301. Revisions to
AWWA C301 were adopted in 1955, 1958, 1964, 1972, 1979, 1984, 1992, and 1999. A revision
in 2007 is anticipated. Those standards were compared in an attempt to determine what effect, if
any, the changes in the standards may have had on the number and frequency of PCCP failures.
55
Table 2.3
Production of prestressed concrete cylinder pipe
PCCP Production ( lin. ft. )
Production by Manufacturer by Year

LockJoint / Interpace /
Year
Total
LCP Total
ECP Total
Ameron
LCP
ECP
CRETEX
LCP
ECP
Hanson
LCP
GHA-LockJoint
ECP
LCP
ECP
Price Brothers
LCP
United Concrete Pipe
ECP
LCP
ECP
1940
56
1942
21,000
21,000
21,000
1943
226,675
226,675
167,680
1945
200,366
200,366
200,366
1946
1,305,314
1,305,314
1947
447,685
447,685
1948
893,852
893,852
1949
725,410
725,410
1950
805,070
805,070
1951
925,834
914,918
1952
931,188
1953
1,900
58,995
1,303,414
304,616
143,069
19,800
859,148
14,904
26,909
39,263
517,834
141,404
6,601
20,407
700,923
77,139
10,916
22,885
752,356
139,677
10,916
930,238
950
44,832
692,576
192,830
950
1,294,097
1,254,119
39,978
260,368
21,312
1954
1,752,670
1,482,035
1955
1,991,826
1956
38,548
56,290
18,666
898,913
270,635
58,721
26,312
997,117
243,473
426,197
850
1,437,237
554,589
15,080
5,520
1,167,829
535,933
254,328
13,136
2,284,547
1,860,798
423,749
46,312
31,785
1,422,345
348,321
392,141
43,643
1957
2,182,540
1,797,252
385,288
25,355
88,446
23,637
1,196,023
348,253
487,428
13,398
1958
2,056,495
1,497,426
559,069
67,976
27,741
15,531
1,037,485
445,186
364,224
19,552
78,800
1959
2,179,924
1,576,372
603,552
56,559
46,187
1,101,059
381,306
418,754
31,433
88,560
1960
2,278,095
1,804,354
473,741
35,737
22,535
1,338,091
323,365
414,801
8,341
119,500
1961
2,862,046
1,710,406
1,151,640
20,191
248,590
1,163,154
846,389
527,061
15,987
56,066
15,725
40,674
Table 2.3 Continued


Year
Total
LCP Total
ECP Total
Ameron
LCP
ECP
CRETEX
LCP
ECP
Hanson
LCP
GHA-LockJoint
ECP
Price Brothers
LCP
ECP
LCP
ECP

LCP
ECP
57
1962
2,515,046
1,375,402
1,139,644
17,680
336,533
27,176
3,918
981,273
760,643
349,273
38,550
1963
2,679,433
1,639,768
1,039,665
40,672
86,995
58,580
10,422
977,661
848,924
562,855
88,624
4,700
1964
3,170,979
1,915,776
1,255,203
76,381
97,187
48,452
1,211,747
853,728
606,842
226,642
50,000
1965
3,023,743
1,418,063
1,605,680
67,790
55,803
80,013
1,221,923
694,707
127,786
742,363
20,807
1966
2,877,632
1,779,514
1,098,118
81,186
90,117
39,727
1,211,712
674,073
477,685
210,832
92,300
1967
2,814,977
2,159,074
655,903
85,355
99,238
112,550
1,352,079
342,828
672,102
38,970
76,200
1968
2,621,965
1,922,658
699,307
87,090
234,275
29,860
1,166,508
460,684
521,875
32,728
88,945
1969
2,392,680
1,586,247
806,433
57,766
104,506
77,466
936,319
447,598
545,422
48,999
174,604
1970
2,448,931
1,734,379
714,552
134,385
140,157
142,975
999,987
324,402
594,235
61,921
50,869
1971
3,563,327
2,433,523
1,129,804
101,931
228,738
260,740
1,688,326
639,822
516,459
87,410
39,901
1972
2,244,698
1,564,241
680,457
73,490
111,722
198,665
977,752
332,757
474,767
40,891
34,654
1973
2,757,025
2,072,785
684,240
25,377
125,117
302,625
25,279
1,224,409
377,599
520,374
84,738
71,507
1974
2,723,351
1,965,691
757,660
13,789
71,353
193,879
162,055
1,238,944
423,782
519,079
100,470
1975
2,363,548
1,692,189
671,359
45,827
111,726
302,330
63,222
730,856
443,537
613,176
32,363
1976
1,860,150
1,301,514
558,636
70,573
101,959
196,295
619,320
187,056
545,755
100,303
34,480
1977
2,576,568
1,763,398
813,170
31,725
219,057
258,942
53,716
924,227
482,007
514,791
58,390
33,713
1978
2,601,117
2,017,024
584,093
5,345
76,152
158,169
77,253
1,075,802
327,880
658,947
75,583
27,225
1979
1,802,973
1,246,353
556,620
19,026
13,822
155,033
169,162
668,366
268,272
361,373
91,094
42,555
14,270
1980
2,482,482
1,653,428
829,054
22,650
133,473
482,039
174,810
515,873
440,501
541,831
68,022
91,035
12,248
1981
1,921,147
1,426,200
494,947
46,402
34,981
239,585
42,552
662,033
240,765
468,462
127,830
9,718
48,819
1982
1,358,447
939,746
418,701
20,400
98,210
105,838
38,966
562,156
165,148
251,352
87,377
1983
1,871,979
1,548,688
323,291
113,31
34,643
245,971
45,820
469,506
205,532
719,892
37,296
12,551
35,655
20,511
4,409
118,76
29,000
Table 2.3 Continued


Year
Total
LCP Total
ECP Total
Ameron
LCP
CRETEX
ECP
LCP
Hanson
ECP
LCP
GHA-LockJoint
ECP
Price Brothers
LCP
ECP
LCP
401,170
216,943
554,689
ECP

LCP
ECP
9
1984
1,774,748
1,179,707
595,041
14,931
117,556
150,000
151,849
108,693
58,917
116,94
1985
1,557,419
1,017,591
539,828
128,335
1986
1,789,172
1,186,724
602,448
152,305
44,283
3,352
188,611
143,305
293,920
77,373
418,115
168,515
302,567
224,846
313,077
66,429
526,797
155,516
58
115,26
1987
1,597,626
1,019,775
577,851
88,788
120,976
221,738
183,405
677,061
190,396
1988
1,471,808
912,607
559,201
53,070
141,488
74,515
119,232
231,873
651,887
199,743
1989
1,689,087
1,168,307
520,780
137,175
233,941
28,756
115,480
61,011
818,886
293,838
1990
1,143,370
838,233
305,137
650
87,143
215,968
54,630
145,246
38,798
476,369
124,566
1991
890,779
719,261
171,518
3,452
2,838
37,234
30,799
194,032
48,227
484,543
89,654
1992
937,130
750,103
187,027
8,246
4,362
177,671
75,982
111,521
5,200
452,665
101,483
1993
519,931
316,326
203,605
152
25,048
3,927
23,664
100,323
267,614
99,203
1994
680,547
428,525
252,022
12,274
42,771
43,979
63,147
28,516
321,660
167,253
1995
829,756
445,457
384,299
2,817
55,547
26,910
16,871
71,323
373,039
283,249
1996
501,878
414,210
87,668
783
2,354
114,574
17,028
22,262
5,109
276,591
63,177
1997
393,734
249,260
144,474
300
17,317
3,770
47,536
11,267
21,993
233,923
57,628
1998
502,632
237,256
265,376
10,500
19,350
49,086
59,232
39,936
48,036
137,734
138,758
1999
723,965
551,866
172,099
4,100
44,906
184,528
30,590
79,521
32,522
283,717
64,081
2000
671,838
430,295
241,543
38,089
10,680
64,005
22,555
56,674
70,984
271,527
137,324
2001
423,576
261,934
161,642
20,473
39,067
12,193
222,867
128,976
2002
402,363
194,951
207,412
66,900
31,682
68,209
162,173
72,303
2003
365,653
228,036
137,617
28,000
19,619
7,017
208,417
102,600
947
1,096
22,300
Table 2.3 Continued


Year
Total
LCP Total
ECP Total
Ameron
LCP
CRETEX
ECP
LCP
Hanson
ECP
GHA-LockJoint
LCP
ECP
18,481
LCP
ECP
56,128
326,177
45,637
12,355
22,744
93,394
72,754
9,687
34,848
23,717,49
8
446,423
344,658
101,765
2005
201,947
105,949
95,998
2006
44,535
9,687
34,848
99,596,74
71,056,90
28,539,84
716,72
3,272,04
1,510,89
635,05
6,338,79
4,161,14
Sums
200
500
LCP
2004
38,266,875
ECP
Price Brothers
13,775,216
LCP
ECP
5,526,26
506,12
1,170,12
59
Analyses described within this study have been grouped as families of production, by
the standard in effect at the time of reported manufacture. It is understood that there are
exceptions to this that were not determinable without examination of individual project
construction documents. For example, pipe could have been made conforming to a pending
standard but could also have been specified to an earlier standard. It is assumed that most pipe
was manufactured at least nominally to the standard in effect at time of production.
Design basis, as noted above, was originally established by the pioneers of PCCP.
Beginning with the tentative AWWA 7B.2-T standard, the design basis was addressed within the
body of the standard.
In 1964, two design bases were incorporated as Appendices A and B. Those are
addressed later in this study. In 1992, the appendices were deleted and an entirely new design
basis was adopted as a separate standard, AWWA C304-92.
The principal revisions to these standards are summarized in Table 2.4 and discussed
below.
Size Range
Size range was initially 16 to 48 inches in 1949. That upper limit was increased to 54
inches in 1952 and reduced to 48 inches (for LC-PCCP) in 1955. In 1979, the upper limit for
LC-PCCP was increased to 60 inches in diameter, where it remains today.
The size of EC-PCCP was initially limited between 24 and 72 inches in 1955. The size
limit was increased to 96 inches in the 1964 revision, with a note allowing larger diameters
subject to the approval of the engineer. The 1972 standard increased the maximum size to 144
inches, and in the 1979 standard, a note recognized that larger sizes had been manufactured. The
largest EC-PCCP pipe manufactured to date is 252 inches (21 feet) for siphons on the Central
Arizona Project. Although seemingly innocuous, the furnishing of pipe sized beyond the limits
of the standard may be relevant to manufacturing pipe with other features beyond those
standards limits.
Design Basis
The 1949 tentative AWWA standard and, presumably, the manufacturers standards
design was based upon both the steel cylinder and the wire to simultaneously reach their elastic
limit at 2-1/4 times the working pressure. On the surface, that implies a factor of safety on yield
(or restated as design working stress of those elements at 0.45 Fy). But because the wire exhibits
no elastic limit, this is not correct. What is the elastic limit of the steel wire? The answer is
found in ASTM A82-34, which was in effect and unchanged until 1952. The wire was required
to meet a tensile strength of 80,000 psi and a yield point of 80 percent of that value. Paragraph
4(c) of that standard stated, The yield point shall be determined by the drop of the beam or halt
in the gauge of the testing machine. In case of no definite drop of the beam or halt in the gauge is
observed until final rupture occurs, the test shall be construed as meeting the requirement for
yield point. In other words, the ductility requirement of 20 percent was specifically waived.
In practice then it is assumed that the design of PCCP to meet ...their respective elastic limits...
was based upon 80 percent of the tensile strength of the wire.
60
Table 2.4
AWWA C301 prestressed concrete cylinder pipe timeline
Standard
Year
Approval date
Status
Effective date
Size range Lined Cylinder type
Size range Embedded Cylinder type
Pressure range
Design basis
Design calculations
Design pressure minimum
Design surge allowance
Design earth load minimum
Supply of pipe
Cement (ASTM C150)
7B.2-T
1949
21-Nov-49
Tentative
C301
1952
13-Jun-52
First Ed.
C301
1955
17-Jun-55
Tentative
C301
1958
26-Jan-58
2nd. Ed. (a)
C301
1964
27-Jan-64
3rd. Ed. (a)
C301
1972
31-Jan-72
Standard
C301
1979
28-Nov-79
Standard
C301
1984
30-Jan-84
Standard
16-48 in.
16-54 in.
16-48 in.
24-144 in.
Appendix A or B
submit if required
Appendix A or B
submit if required
16-60 in.
24-144 in. (l)
note n
Appendix A or B
submit if required
16-60 in.
24-144 in. (l)
100 to 600 ft.

note s
16-48 in.
24-72 in.
50 to 350 psi.
note d
submit if required
16-48 in. (i)

24-96 in. (i)
100 to 600 ft.

note b
16-48 in.
24-72 in.
50 to 350 psi.
note d
submit if required
40 psi
40 psi
50 psi (e)
50 psi (e)
40%
40%
6 ft. cover & ordinary 6 ft. cover & ordinary 6 ft. cover & ordinary 6 ft. cover & ordinary
trench, 3 ft AASHO
trench, 3 ft AASHO
trench, AASHO H-20 trench, AASHO H20
"crushing strength"
"crushing strength"
live load
live load
No pipe manufactured No pipe manufactured No pipe manufactured No pipe manufactured No pipe manufactured supply from inventory
until drawings approved until drawings approved until drawings approved until drawings approved until drawings approved
OK
Type II
Type I or II
Type I or II
Type I or II
Type I or II
Type I or II
Fine aggregate organics/impurities
Fig 2, ASTM C40-33
Fig 2, ASTM C40
Fine aggregate quality

Fine aggregate gradation, % passing
No. 200
Fineness Modulus variation
Coarse aggregate gradation
Coarse aggregate deleterious
substances, max %
equal to Ottawa sand

5

5
Admixtures
Steel Cylinders
Cylinder thickness minimum
+/- 0.20
note c
5
ASTM A245-48T
16 ga.
"no darker than ref.

std." in ASTM C40
5

std." in ASTM C40
5

std." in ASTM C40
5
ASTM C33-71a Sec.

4.2
gradation in Table 1
5
Appendix A or B ( r)
submit if required
none (f)
none (f)
none (f)
ASTM C33
ASTM C33, Class 3S,
3M, or 1N in Table 3,
Fig 1.
ASTM C494
Fy = 33 ksi min. ASTM
A569, ASTM A570,
ASTM A611 gr. C, or
D, AISI Manual 1214
18 ga. < 48in., 16 ga < 18 ga. < 60 in., 16 ga < 18 ga. < 36 in., 16 ga >
0.0598 in. (16 ga.)
36 in.
54 in
60 in
Mill tests of steel
submit in advance of mfrsubmit in advance of mfrsubmit in advance of mfrsubmit in advance of mfr
made available
made available
mill tests or plant tests mill tests or plant tests mill tests or plant tests
at mfr's option made
available
available
available
Hydrostatic test w/ joint rings attached 20 ksi min, 25 ksi max. 20 ksi min, 25 ksi max. 20 ksi min, 25 ksi max. 20 ksi min, 25 ksi max. 20 ksi min, 25 ksi max. 20 ksi min, 25 ksi max. 20 ksi min, 25 ksi max. 20 ksi min, 25 ksi max. 20 ksi min, 25 ksi max.
(p)
(p)
(p)
Bend and tensile tests of cylinder
per AWWA 7A.3-1940
per AWWA C201
per AWWA C201
per AWWA C201
none
Wire reinforcement
ASTM A229-41,
ASTM A227, or ASTM
ASTM A227
ASTM A227
ASTM A227 (h)
ASTM A227 (h)
ASTM A648 (h)
ASTM A648
ASTM A648
ASTM A227-47T, or
A82
ASTM A82-34
Wire size minimum
1/8 in. dia.
1/8 in. dia.
6 gauge
6 gauge (0.192")
8 gauge (0.162")
0.162"
0.162"
0.162"
0.192 in. (6 ga.)
Wire anchorage strength
"equivalent to strength "equivalent to strength "equivalent to strength "equivalent to strength
"minimum ultimate
75% Fu
75% Fu
75% Fu
75% Fu
of wire"
of wire"
of wire"
of wire"
tensile strength of wire"
Wire wrapping stress
0.7Fu
0.7Fu
0.7Fu
0.75Fu
0.75Fu
0.75Fu
0.75Fu
0.75Fu
Maximum wire spacing
1 in.
1.0", except for no. 6 & 1.5" for no. 6 wire, 1.0" 1.5" for no. 6 wire, 1.0" 1.5"; 1" for LC with >
1.5"; 1" for LC with
larger, 1.5" if < 100psi
for larger wire
for larger wire
No. 6 wire (j)
wire > 0.192"
wire > 0.192"
wire > 0.250"
wire > 0.250"
Minimum wire spacing
0.3 in.
0.3 in.
3/16 in. clear
3/16 in. clear
3/16 in. clear
3/16 in. clear
3/16 in. clear
3/16 in. clear
2.75 wire dia. for LC &
2.0 wire dia.for EC
Min. core thickness
Table 3
Table 3
Table 3
Table 3a
Table 3
1/16 ID
1/16 ID
1/16 ID
1/16 ID
Core out of roundness
"round and true"
"round and true"
"round and true"
"round and true"
"round and true"
"round and true"
1% of average
1% of average
0.7% for LC, 0.5% for
EC
61
C301
1999
24-Jan-99
Standard
1-Sep-99
16-60 in.
24-144 in. (l)
C301
2007
pending
16-60 in.
24-144 in. (l)
AWWA C304
AWWA C304
AWWA C304
submit "verification of submit "verification of submit "verification of
the design"
the design"
the design"
AWWA C304
AWWA C304
AWWA C304
40 psi
40 psi
40%
40%
6 ft. cover & ordinary 6 ft. cover & ordinary
trench, AASHTO H20 trench, AASHTO H20
live load
live load
supply from inventory supply from inventory supply from inventory
OK
OK
OK
Type I or II
Type I or II, and fly ash, Type I or II, and fly ash,
pozzolans up to 20% pozzolans between 10%
substitution for cement & 20% substitution for
cement
ASTM C33-77 Sec. 5
ASTM C33
ASTM C33 except for
sections 7.2 & 7.3
ASTM C33-71a Sec.

4.2
+/- 0.20
+/- 0.20
+/- 0.20
+/- 0.20
+/- 0.20
+/- 0.20
ASTM C33
note c
note c
note c
note g
note g
note g
5
5
5
5
5
ASTM C33 Table 3
class 4S exc. soundness
test
ASTM C494
ASTM C494
ASTM C494
ASTM A245, gr. B or C ASTM A245, gr. B or C ASTM A245 gr. B or C ASTM A415, max c ASTM A570 gr. B or C, ASTM A570, or ASTM Fy = 30 ksi min. ASTM
0.25%, Fy = 27 ksi min. or ASTM A569, max c A569, max c 0.25%, A570, ASTM A611 gr.
0.25%, Fy = 27 ksi min. Fy = 27 ksi min., or
B, C, or D, AISI
ASTM A611, gr. B, C,
G10120 through
D, or E (m)
G10200, or ASTM
A569, max c 0.25%,
16 ga.
C301
1992
22-Jun-92
Standard
1-Apr-93
16-60 in.
24-144 in. (l)
supply from inventory

OK
Type I or II, and silica
fume, fly ash, pozzolans
between 10% & 20%
substitution for cement
ASTM C33
supply from inventory

OK
Type I or II. Fly ash,
pozzolans up to 20%
substitution for cement,
silica fume up to 10%
ASTM C33
ASTM C33
ASTM C33, Class 3S,
Fig 1.
ASTM C494
Fy = 33 ksi min. ASTM
A569, ASTM A570,
ASTM A611 gr. C, or
D, ASTM A635 gr.
1012 through 1020,
ASTM A659, or ASTM
A907
0.0598 in. (16 ga.)
ASTM C33
ASTM C33, Class 3S,
Fig 1.
ASTM C494
ASTM A1011 SS,
ASTM A635, ASTM
A569, or ASTM A1018
SS
0.0598 in. (16 ga.)
mill tests or plant tests mill tests or plant tests

available
available
20 ksi min, 25 ksi max. 20 ksi min, 25 ksi max.
(p)
(p)
ASTM A648
ASTM A648 including

the supplementary
requirements
nominal 0.192" (6 ga.) nominal 0.192" (6 ga.)
75% Fu
75% Fu
0.75Fu
0.75Fu
wire > 0.250"
wire > 0.250"
2.75 wire dia. for LC & 2.75 wire dia. for LC &
2.0 wire dia.for EC
2.0 wire dia.for EC
1/16 ID
1/16 ID
greater of 3/16" or 0.7% greater of 3/16" or 0.7%
for < 48 in. dia., 1/2" or for < 48 in. dia., 1/2" or
0.5% for pipe > 4in.
0.5% for pipe > 4in.
(continued)
Table 2.4 (Continued)

Standard
Core ID tolerance
Rubber gaskets
65% natural rubber
50% natural or
synthetic
50% natural or
synthetic
50% natural or
synthetic
50% natural or
synthetic
C301
1/4" for < 36", 3/8" for
42 & 48", 1/2" for >
54", and 3/4" for > 84
in.
1/8" for < 36", 3/16" for
42 & 48", 1/4" for 54"
to 72" & 3/8" for > 72"
3/16 in. for < 36", 1/4"
for larger
3/16 in. for gaskets
21/32 in. dia. or less,
and 1/4" for greater
dia.(k)
50% natural or
synthetic
Gasket diametral tolerance

Core cement content/cu. yd.
Core water/cement ratio
7 sacks
"to meet strength"
7 bags
"to meet strength"
7.0 bags
"to meet strength"
7 sacks
"to meet strength"
6 sacks
"to meet strength"
+/- 1/64 in.

6 bags
"to meet strength"
+/- 1/64 in.

560 lb
0.6 max.
+/- 1/64 in.

560 lb
0.6 max.
C301
1/4" for < 36", 3/8" for
42 & 48", 1/2" for >
54", and 3/4" for > 84
in.
1/8" for < 36", 3/16" for
42 & 48", 1/4" for 54"
to 72" & 3/8" for > 72"
3/16 in. for < 36", 1/4"
for larger
dia.(k)
50% synthetic
polyisoprene or other
synthetic rubbers
+/- 1/64 in.
560 lb
0.5 max.
7 day compressive strength

28 day compressive strength
Wire wrap core after placing
2,600 psi, min.

4,500 psi, min.
7 days
2,600 psi, min.

4,500 psi, min.
7 days
3,000 psi, min.

4,500 psi, min.
core reaches 7 day
strength
0.55 fcw
3,000 psi, min.

4,500 psi, min.
core reaches 7 day
strength
0.55 fcw
3,000 psi, min.

4,500 psi, min.
core reaches 7 day
strength
0.55 fcw
4,500 psi, min.

core reaches 3,000 psi
0.40 fcw
2,600 psi, min.

4,500 psi, min.
core reaches 7 day
strength
0.40 fcw
3,000 psi, min.

4,500 psi, min.
0.40 fcw
2,600 psi, min.

4,500 psi, min.
core reaches 7 day
strength
0.40 fcw
36 hr. min.
36 hr. min.
32 hr. min.
1 cement to 3 sand
1 cement to 3 sand
1 cement to 3 sand
3/4 in. min.
3/4 in. min.
7/8 in.
Core thickness tolerance
7B.2-T
1/4 in.
+/- 1/8 in.
Joint rings - min. thickness of
3/16 in.
Joint ring tolerance
1/32 in.
Core stress induced at time of wrap

Slurry under wire
Steam or water curing
Mortar coating mix
Soluble chloride content
Mortar coating thickness
Mortar strength (cubes, ASTM C109)
Concrete coating mix
Soluble chloride content - conc. ctg.
Concrete coating thickness
Coating compressive strength, 7 day
Coating compressive strength, 28 day
Coating steam cure
Coating water cure
Seal coat
Openings allowed in pipe wall
C301
C301
C301
C301
1/4" for < 36", 3/8" for 1/4" for < 36", 3/8" for 1/4" for < 36", 3/8" for 1/4" for < 36", 3/8" for
42 & larger
42 & 48", 1/2" for >
42 & 48", 1/2" for >
42 & 48", 1/2" for >
54"
54"
54", and 3/4" for > 84
in.
+/- 1/8 in.
1/8" for < 36", 3/16" for 1/8" for < 36", 3/16" for 1/8" for < 36", 3/16" for
42 & 48", 1/4" for >
42 & 48", 1/4" for > 42 & 48", 1/4" for 54"
to 72" & 3/8" for > 72"
54"
54"
3/16 in.
3/16 in. for < 36", 1/4" 3/16 in. for < 36", 1/4" 3/16 in. for < 36", 1/4"
for larger
for larger
for larger
1/32 in.
1/32 in.
1/32 in.
3/16 in.
1 cement to 4 aggregate 1 cement to 4 aggregate
7 bags
1 in. min.
1 in. min.
1 1/2 in.
2,600 psi, min.

4,500 psi, min.
12 hr. min.
2,600 psi, min.

4,500 psi, min.
12 hr. min.
2,600 psi, min.

4,500 psi, min.
12 hr. min.
4 days min.
4 days min.
4 days min.
C301
1/4" for < 36", 3/8" for
42 & 48", 1/2" for >
54", and 3/4" for > 84
in.
1/8" for < 36", 3/16" for
42 & 48", 1/4" for 54"
to 72" & 3/8" for > 72"
3/16 in. for < 36", 1/4"
for larger
dia.(k)
50% natural or
synthetic
C301
1/4" for < 36", 3/8" for
42 & 48", 1/2" for >
54", and 3/4" for > 84
in.
1/8" for < 36", 3/16" for
42 & 48", 1/4" for 54"
to 72" & 3/8" for > 72"
3/16 in. for < 36", 1/4"
for larger
dia.(k)
50% natural or
synthetic
C301
1/4" for < 36", 3/8" for
42 & 48", 1/2" for >
54", and 3/4" for > 84
in.
1/8" for < 36", 3/16" for
42 & 48", 1/4" for 54"
to 72" & 3/8" for > 72"
3/16 in. for < 36", 1/4"
for larger
dia.(k)
50% synthetic
synthetic rubbers
+/- 1/64 in.
560 lb
0.5 for pipe centrifigal
cast, 0.45 for vert. cast
or radial compaction.
C301
1/4" for < 36", 3/8" for
42 & 48", 1/2" for >
54", and 3/4" for > 84
in.
1/8" for < 36", 3/16" for
42 & 48", 1/4" for 54"
to 72" & 3/8" for > 72"
3/16 in. for < 36", 1/4"
for larger
dia.(k)
50% synthetic
synthetic rubbers
+/- 1/64 in.
560 lb
0.5 for pipe centrifigal
cast, 0.45 for vert. cast
or radial compaction.
4,500 psi, min.

4,500 psi, min.

0.55 fcw
0.55 fcw
0.55 fcw
0.55 fcw
1 sack/10 gal. water
94 lbs./10 gal. water
32 hr. min.
32 hr. min.
24 hr. min.steam, 32 hr. steam: note q, water: 12 steam: note q, water: 12 steam: note q, water: 12 steam: note t, water: 12 steam: note t, water: 12
min. water
hr. min.
hr. min.
hr. min.
hr. min.
hr. min.
1 cement to 3 sand
1 cement to 3 sand
1 cement to 3 sand
1 cement to 3 sand
1 cement to 3 sand
1 cement to 3 sand
1 cement to 3 sand
1 cement to 3 sand
0.06%, max.
0.06%, max.
0.06%, max.
0.06%, max.
0.06%, max.
7/8" nom., 3/4 in. min.
5/8" min. over wire
5/8" min. over wire
5/8" min. over wire
3/4" min. over wire
3/4" min. over wire
3/4" min. over wire
3/4" min. over wire
5,500 psi. min.
5,500 psi. min.
5,500 psi. min.
7 bags
7 bags
7 bags
660 lbs.
deleted
0.06%, max.
1.5" nom., 1" min.
1.5" nom., 1" min. over 1.5" nom., 1" min. over 1.5" nom., 1" min. over
core
core
core
2,600 psi, min.
3,000 psi, min.
3,000 psi, min.
3,000 psi, min.
4,500 psi, min.
4,500 psi, min.
4,500 psi, min.
4,500 psi, min.
12 hr. min.
12 hr. min.
12 hr. min.
note q, 12 hr. min, after note q, 12 hr. min, after
12 hr. min.
12 hr. min.
12 hr. min.
6 hr. water cure OK
6 hr. water cure OK
4 days min.
4 days min.
4 days min.
4 days min.
4 days min.
4 days min.
4 days min.
4 days min.
AWWA C104 if ordered AWWA C104 if ordered
Sec. 4.4
Sec. 4.4
Sec. 4.7.4
Sec. 4.7.4
Notes
Red indicates a change from previous version of the standard.
(a) All printings of the previous edition are obsolete.
(b) "the steel cylinder and wire shall reach their respective elastic limits simultaneously at a pressure equivalent to at least 2 1/4 times the normal water pressure"
(c) Subject to the approval of the engineer.
(d) "At a pressure equal to twice the design pressure, the stress shall not exceed its original gross wrapping stress." Combination dead & live load < 110% incipient cracking of core.
(e) "not a specific water hammer allownacebut results fromn the conservative factors used in the design"
(f) No minimum stated, only implied 16 ga. based upon section 3.2.
(g) "...well graded subject to approval of the engineer"
(h) "except that wire with higher tensile strengths [exceeding class {II}[III] ] may be used if it meets the other requirements of ASTM A227."
(i) Note below table 3 indicates "For embedded cylinder pipe larger than 96 in. diameter, dimensions and details of design shall be subject to approval"
(j) Note no maximum spacing for No. 8 wire.
(k) Also, out of roundness < 0.5% of average ID.
(l) "Larger sizes have been manufactured"
(m) "Alternatively, hot-and cold-rolled steel sheets ansd strips having a minimum yield strength of 27,000 psi and conforming to AISI 1012 through 1020 specifications shall be acceptable."
(n) For minimum core & cylinder, 250 psi for 16-20 in. LC, 200 psi for 24-36" LC, 175psi for 42" pipe, & 100 psi for 54 & 60" pipe. EC is unlimited.
(p) Cylinders > 10 ga. shall be tested to same pressure as 10 ga.
(q) "sufficient to produce the concrete strengths specified"
( r) Advisory included in foreword stating that the standard "does not contain all of the information needed" and also recommends use of AWWA Manual M9 to supplement the standard.
(s) "such that the core will be sufficiently compressed to withstand an internal hydrostatic pressure equal to at least 1.25 times the designed pressure without tensile stress being induced in the core."
(t) "concrete curing compound suitable " allowed, then "sufficient to produce the concrete strengths specified"
62
Also, the tentative standard did not include a limit on the decompression stress of the
concrete core. That was rectified in the 1952 standard with the requirement that the core will
be sufficiently compressed to withstand an internal hydrostatic pressure equal to at least 1.25
times the design pressure without tensile stress being introduced in the core. On the surface,
that would imply an increased factor of safety, but the allowable stress at water pressure was also
increased to 70 percent of ultimate. The overall result was a lowering of the ultimate safety
factor for the affected components.
The 1955 standard incorporated a minimum 50-psi design allowance for surge pressures
and also addressed submittal of the design if required by the project specifications.
The 1964 revision, as indicated previously, moved the design basis to the appendices.
Design according to Method A used a semi-empirical approach based on Wo, which is ninetenths of the three-edge bearing test load that causes incipient cracking and the theoretical
internal pressure (Po) that relieved the calculated residual compression in the concrete core due
to prestressing exclusive of external load. The allowable combinations of three-edge bearing
load and internal pressure were determined by a cubic parabola, passing through Wo and Po,
which defined the limit of such combinations. The three-edge bearing loads used in Method A
were converted to earth loads and transient external loads using bedding factors such as
described in AWWA Manual M9, Concrete Pressure Pipe (1979) and ACPA, Concrete Pipe
Design Manual (1988). That procedure also allowed a 20 percent increase on the external load
and 40 percent increase on the core decompression load for short-duration (i.e., surge) events.
Appendix B was similar to Appendix A except its axes would be determined by analysis
and the intercepts determined as straight lines. The Method B procedure limited the maximum
combined net tensile concrete stress in pipe under static external load and internal pressure to a
value equal to 7.5 f ' c , where fc equals the 28-day compressive strength of core concrete in
psi. Coefficients for moments and thrusts to determine stresses were to be per Paris (1921) or
Olander (1950).
Both design methods limited the working pressure for embedded-cylinder pipe to Po and
to 0.8Po for lined-cylinder pipe. Under transient conditions, such as those produced by surge
pressures and live loads, both methods permitted the above-noted increased allowable internal
pressure and external load.
The 1964 revision also included a minimum design pressure of 40 psi.
No substantial revision occurred to the design basis (other than wire stresses discussed
separately) until the 1992 elimination of the design appendices and the adoption of the design
standard AWWA C304. The 1992 design standard was a significant change in design approach
to a strength and serviceability basis. The development of that design standard is described in the
foreword to the standard and elsewhere (Heger, Zarghamee, and Dana 1990; Zarghamee 1990a;
Zarghamee and Fok 1990b; Zarghamee, Fok, and Sikiotis 1990c; Zarghamee and Dana 1991).
Concrete Quality
The 1949 tentative standard only included Type II cement. In 1952, Type I was allowed
and it stayed that way until fly ash and other pozzolans were allowed as a substitute for up to 20
percent of the cement in 1984. That substitution was revised in 1992 to between 10 and 20
percent. In 1999, silica fume was allowed on equal basis with fly ash and other pozzolans but
was limited in the pending 2007 revision to 10 percent. The effect of fly ash, silica fume, and
63
pozzolans is to increase the density and reduce the permeability of the concrete. Admixtures as
described in ASTM C494 have been allowed since the 1972 standard.
Aggregate quality, as defined by the impurities allowed in the fine and coarse aggregate,
size, and gradation, has varied with each standard revision. Since the 1972 edition, some
modification or outright adoption of ASTM C33 has occurred. For more details, refer to
Table 2.4. There were no statistically significant data to indicate that these changes in specified
materials affected the performance of PCCP. The forensic examination of individual specimens
may ultimately indicate otherwise. The effects of creep, shrinkage, and temperature on concrete
are a result of the quality of the concrete, which were addressed in the new design standard,
AWWA C304-92. In order to accommodate the new design standard, for the purpose of
qualifying a source of aggregate to be used in the manufacture of pipe, the apparent specific
gravity of fine aggregate was required to be sampled in accordance with ASTM D75 and
measured in accordance with ASTM C128 when not less than 2.6. Specific gravity
measurements and all other mandatory tests listed in ASTM C33 also had to be made at sixmonth intervals or whenever the source of fine aggregate is changed. Sieve analysis of fine
aggregate also was required to be performed in accordance with ASTM C33 on a weekly basis
(Section 2.2).
For the purpose of qualifying a source of coarse aggregate to be used in the manufacture
of pipe, the 1992 standard also required the apparent specific gravity of coarse aggregate, when
sampled in accordance with ASTM D75 and measured in accordance with ASTM C127, to be
not less than 2.6. Specific gravity measurements and all other mandatory tests listed in ASTM
C33 had to be made at six-month intervals or whenever the source of coarse aggregate is
changed. Sieve analysis of coarse aggregate also was required to be performed in accordance
with ASTM C33 on a weekly basis. Never before had deleterious materials been addressed, but
the 1992 standard also required coarse aggregate to meet the deleterious substances requirements
for Class 3S, 3M, or 1N shown in Table 3 of ASTM C33 in the severe, moderate, and negligible
weathering regions, respectively, shown in Figure 1 of ASTM C33.
Steel for Cylinders
Many different ASTM standards have been referenced in the PCCP standards, likely
because of changes to material availability and changing ASTM standards. There were no
statistically significant data to indicate that these changes materially affected PCCP performance.
For more details, refer to Table 2.4.
Steel Cylinder Thickness
The requirements for steel cylinder thickness changed significantly since the original
development of PCCP. They are illustrated in Figure 2.6. The original design basis was for 16gauge (0.060-inch thick) steel, and that was incorporated in the (tentative) 1949 and 1952
standards. The 1955, 1958, and 1964 standards did not address minimum cylinder thickness.
Rather, it was implied in Section 3.2, Maximum design pressures for the various core
thicknesses and 16-gauge cylinders shall be those indicated in Tables 3a and 3b (1955 and
1958) and These values are predicated on the use of 16-gauge cylinders and the concrete
strength specified in Sec. 3.6 (1964). The 1972 standard allowed 18-gauge (0.048-inch thick)
steel for PCCP 48 inches and 16 gauge for 54 inches and greater. The 1979 standard increased
64
the diameter for which 18-gauge cylinders were allowed to 60 inches but that was reduced to 36
inches in the 1984 standard. The 1992 standard and subsequent revisions have required 16-gauge
minimum for all sizes.
Steel Cylinder Weld Tests
The 1949 through 1958 standards required bend and tensile tests to be conducted on the
cylinder weld seams. These requirements were omitted from later standards. Those test
requirements referenced the comparable AWWA steel pipe weld seam tests. It is probable that
these tests were omitted because of the difficulty in getting lap seams (how many of the pipe
cylinders were made) to pass bend tests. Currently, the only test of those welds is the hydrostatic
shop test of the cylinder with joint rings attached.
Wire Reinforcement
The type of wire used in PCCP varied widely. It is discussed separately below.
The minimum size (diameter) of reinforcement changed significantly with each standard
revision. The variation with time in the minimum wire size is illustrated on Figure 2.7. Initially,
1/8 inch (0.125 inch) in the 1949 and 1952 standards, the minimum size was increased 235
percent by volume to 6 gauge (0.192 inch) in the 1955 standard then reduced to 8 gauge (0.162
inch) in the 1964 standard. The 1992 standard increased the minimum again to 6 gauge (0.192
inch) where it remains today. The use of 8-gauge wire reinforcement (1964 to 1984) may be the
most problematic change in PCCP manufacture. This is discussed separately below.
Wire Wrapping Stress
Research of records to date has not revealed conclusive evidence of what stress the wire
reinforcement was wrapped on the cylinder (for LC-PCCP) or concrete core (for EC-PCCP),
prior to the 1952 standard. In that year, the allowable wire stress was 70 percent of ultimate
tensile wire strength (0.7 Fy). In the 1964 standard the wire stress allowable was increased to 75
percent of ultimate strength (0.75 Fy) where it remains today. The anchorage of the wire was
initially required to have strength equivalent to the strength of the wire, but the 1992 standard
reduced that to the same strength value as the wire at wrap. That also is unchanged today.
Wire Reinforcement Spacing
Wire reinforcement spacing varied from standard revision to standard revision. The
interested reader is referred to the timeline table for more details, but what is important is the
minimum clear distance between wires so as to provide adequate room for placement of the
cement mortar coating. Table 2.5 shows the minimum wire clearance for the minimum wire size.
Most significant is the 66 percent increase in minimum wire clearance in the 1984 standard,
providing adequate clearance for mortar to fully protect the wire.
65
Table 2.5
Minimum wire clearance for minimum wire size
Minimum Clearance
Standard
(years)
1949
Maximum Clearance
LC-PCCP
(inches)
EC-PCCP
(inches)
LC-PCCP
(inches)
EC-PCCP
(inches)
0.175
0.875
1952
0.175
1.375
1955
0.188
0.188
1.308
1958
0.188
0.188
1.308
1964
0.188
0.188
1.338
1972
0.188
0.188
1.338
1979
0.188
0.188
1.338
1984
0.188
0.188
(1)
1.308
1.308
(2)
1.338
1.338
1.338
1.338
(3)
1.338
(3)
1.308
1992
0.336
0.192
1.308
1999
0.336
0.192
1.308
1.308
2007 Pending
0.336
0.192
1.308
1.308
(1)
If 100 psi design working pressure.
(2)
0.838 inch for LC with No. 6 wire or more.
(3)
0.750 inch for LC with 1/4-inch wire or more.
It is curious that the maximum wire spacing is reduced with larger sized wire
reinforcement.
Concrete Core Manufacturing Tolerances
Until the 1979 standard, the core was required to be round and true. In the 1979
standard, the out-of-roundness was allowed to be 1 percent of the measured difference between
the maximum and minimum diameters. That large tolerance coupled with the joint ring
tolerances discussed below may have had a significant effect on PCCP performance. Other
tolerances in the core dimensions (internal diameter and thickness) may not have individually
affected the PCCP performance but tolerances are cumulative. See the timeline (Table 2.4) for
more details.
Joint Ring Tolerances
Until the 1958 standard, joint rings had to meet a tolerance of 1/32 inch (0.031 inch) on
the design clearance. That meant that the nominal space between a round and true bell-andspigot would be zero. In actual practice, the cumulative tolerances for a 42-inch LC-PCCP made
to the standard tolerances could have joint dimensions as shown in Table 2.6.
66
Table 2.6
Joint dimensions
Tolerance
+
Nominal ID
42.0
42.0
Core ID
42.375
41.625
Core thickness at spigot
42.563
41.437
Difference: worst case within tolerance

Tolerance on diameter not on ovality
Worst-case interference
Note
Round and true
1.126 inches
0.031 inch
1/32 inch
1.095 inches
In actual practice, of course, the joint rings were separately made, so that based upon
circumference, the nominal fit was likely to meet the 1/32-inch clearance tolerance. But also in
actual practice, the joint rings were cast within the cores made to the tolerances of the standard.
Because there is no ovality tolerance, the mismatch of joints is likely the cause of many joint
leaks.
Strength of Concrete
Factors that affect the strength of concrete include the amount of cement per unit volume
and the water-cement ratio. The minimum seven-day strength in the standard up until the 1964
standard was 2,600 psi, when it was increased to 3,000 psi. The 1992 standard and subsequent
standards do not address minimum strength at seven days. All the standards require 4,500-psi
minimum at 28 days.
The amount of cement in the core required per cubic yard was seven 94-pound sacks until
1964, when six 94-pound sacks were required. It is curious that this decrease in required cement
content accompanied an increase in required early strength. The 1979 and subsequent standards
slightly reduced the minimum cement content to 560 pounds per cubic yard.
From 1949 until the 1979 standard, the only prescription regarding water-cement ratio
was to assure that the concrete will meet the strength requirements. But in 1979 and 1984,
the standard put a limit of 0.6 on the water-cement ratio. In 1992, that was further reduced to 0.5
and in 1999, for vertically cast or radially compacted cores, that was further reduced to 0.45.
Core Prestressing Limits
Limitations on when the core could be wrapped and the stresses induced by wrapping
have been addressed in each of the standards. The intent is to avoid cracking the core as it is
squeezed at one end during wrapping and unstressed at the other. Another intent of these limits is
to avoid crushing of the core. To that extent then the 1949 and 1952 standards limited winding of
wire around the core until at least seven days after the concrete lining is placed.
In 1955, that absolute time was negated by a performance level or until the concrete
has reached the minimum seven-day compressive strength specified. The net effect was to
speed up pipe production.
Those time limits were coupled with a limit on the design stress induced in the core of 40
percent of the compressive strength at the time of wrap. In other words, a concrete core at seven
67
days age meeting the minimum 2,600-psi compressive strength could only be wrapped such that
the compressive stress induced would not exceed 1,040 psi (0.4 x 2,600 psi).
The 1964 standard increased the compressive stress allowable to 55 percent of the core
concrete strength at time of wrap (fcw). That minimum strength was also increased to 3,000 psi
in 1964, so the induced compressive stress could be as much as 1,650 psi at the minimum
specified compressive stress. There was (and is) no upper limit in the standards on the design
core concrete compressive strength, although practical limits exist.
The seven-day compressive strength requirement was deleted from the standard in 1992.
The effect of this deletion was to allow wrapping of wire on cores younger than seven days
which had reached the 3,000-psi minimum core strength. The imposed core stress at time of
wrapping remained at 0.50 x fcw.
An improvement added to the 1984 standard was the requirement that a thin cement
slurry be placed immediately prior to wrapping the wire. That was intended to ensure complete
encapsulation of the wire in the high pH portland cement environment. Some applications have
been discovered to have left a void beneath one quadrant of the wire, rather than all of the wire
resting on a bed of mortar.
Concrete Coating
Initial concrete coating was cast around the wire-wrapped core, likely using the same
forms used for other concrete pipe products. The minimum cast coating thickness over the core
has varied from 1 inch minimum in 1949 to 1.5 inches in the 1955 standard to 1.5 inches
nominal and 1 inch minimum in 1958. That is illustrated on Figure 2.8. (The minimum wire size
is also graphically represented.) The 1964 standard reduced that cover to 1 inch over the core so
that the minimum cover over the wire could have been 5/8 inch using 3/8-inch wire. The cast
coating was deleted in 1984.
Sprayed, shotcrete-type mortar coating was allowed in the 1949 standard at minimum
thickness of 3/4 inch. The thickness was increased to 7/8 inch in the 1955 standard. In 1958, the
7/8-inch thickness became nominal with a 3/4-inch minimum standard. In 1964, the minimum
coating was reduced further to 5/8 inch. That proved to be inadequate and the 3/4-inch coating
was restored in the 1984 standard.
Summary
It can be seen in examining Table 2.4 and Figure 2.2 that the initial design basis for
manufacture of PCCP appeared to be conservative, and as experience was gained and
competitiveness with other pipe materials increased, that changes were made to reduce the unit
cost of manufacture. In sum, those changes tended to increase the stress level in the pipe at
working pressures and reduced the margin for error.
The most negatively influential year in terms of changes to the standard was 1964
because:
Design Appendices A and B were adopted, moving the design criteria from the body
of the standard to an appendix.
No minimum cylinder thickness (since 1955 edition) was included.
18.5 percent reduction in minimum wire size occurred.
68
The standard allowed a 37.5 percent increase in concrete core stress at time wire is
wrapped.
The standard allowed a 16.7 percent reduction in the minimum amount of portland
cement in the core.
The standard allowed a 20 percent reduction in the minimum coating thickness.
The trend toward reducing conservatism of the product through revisions in the standard
began to reverse course in 1984 with the issuance of AWWA C301-84. That year saw the
allowable additions of fly ash and other pozzolans in an attempt to increase the density of the
concrete coating and core, the incorporation of ASTM C33 for concrete and mortar aggregate
requirements, the slurry placement under the wire, and the minimum coating thickness increased
to 3/4 inch. Significant revisions to the standard in 1992 and adoption of the very detailed design
standard C304-92 appear to have resulted in much improved performance of as-installed PCCP.
CHANGES IN MATERIAL STANDARDS
By far the most influential material item in the performance of PCCP pipe is the wire.
The wire is the means of effecting the prestress, and without it maintaining the core in
compression, the pipes service life is compromised. In order to examine the effect of changing
standards in the wire and its use within the AWWA PCCP standards, a separate timeline was
prepared listing the ASTM standards and comparing some of the important requirements within
those standards. Table 2.7 compares chemical and mechanical requirements, for the cited
standards in effect at the time they were referenced.
Table 2.7
Timeline of wire standards for PCCP
Item
ASTM A82-34
Incorporated in ASTM Book of Standards
1934
ASTM A229-41
ASTM A227-47
ASTM A227-63
1941
1941, rev. 1947
1964
0.55 to 0.75
0.65 to 0.75 (a)
0.45 to 0.75 (a)
0.8 to 1.2 (n)
0.6 to 1.20 (b)
0.6 to 1.20 (b)
Phosphorous, maximum, per cent
0.045
0.045
0.040
Sulfur, maximum, per cent
0.05
0.05
0.050
0.10 to 0.30
0.10 to 0.30
0.10 to 0.30
1 diameter
1 diameter
1 diameter
2 diameters
2 diameters
2 diameters
Carbon, per cent

Carbon, per cent, Class I
Carbon, per cent, Class II
Carbon, per cent, Class III
Manganese, percent
Manganese, per cent, Class I
Manganese, per cent, Class II
Manganese, per cent, Class III
Silicon, per cent

Nitrogen, maximum, percent
Wrap test, 0.162" & smaller
Wrap test, 0.02" < 0.162" & smaller, Class I
Wrap test, 0.02" < 0.162" & smaller, Class II
Wrap test, < 0.312" (0.3 in. in ASTM A 82-34)
1 diameter
Wrap test, 0.162" < 0.312", Class I

Wrap test, 0.162" < 0.312", Class II
(continued)
69

Item
ASTM A82-34
ASTM A229-41
ASTM A227-47
ASTM A227-63
2 diameters
not addressed
not addressed
"not applicable"
ASTM A82-34
ASTM A229-41
ASTM A227-47
ASTM A227-63
80 min.
200 min, 230 max.
200 min, 230 max.
200 min, 230 max.
80 min.
192 min, 220 max.
192 min, 221 max.
192 min, 221 max.
ASTM A227-64
ASTM A227-68
ASTM A227-71
ASTM A648-72
1965
1971
1972
1973
Wrap test, 0.162" < 0.312", Class III

Wrap test, 0.312", Class III
Wrap test, > 0.312" (0.3 in. in ASTM A 82-34)
Reduction of area test (ASTM A370, Sup. IV)
Reduction of area test, 0.192" wire
Item
Reduction of area test, > 0.192" wire
Relaxation Test (ASTM E328)
Torsion test
Splitting Test
Tensile strength, ksi, 8 gauge (0.162")
Tensile strength, ksi, 8 gauge, Class I
Tensile strength, ksi, 8 gauge, Class II
Tensile strength, ksi, 8 gauge, Class III
Tensile strength, ksi, 6 gauge , Class I
Hydrogen embrittlement test (ASTM A1032)
Item
Carbon, per cent
0.45 to 0.75 (a)
0.45 to 0.85 (c)
0.45 to 0.85 (c)
0.45 to 0.75 (c)
0.60 to 1.20 (a)
0.45 to 0.85 (c)
0.45 to 0.85 (c)
0.50 to 0.85 (c)
0.55 to 0.93 (c)
Manganese, percent
0.6 to 1.20 (b)
0.6 to 1.20 (b)
0.6 to 1.30 (b)
0.60 to 1.10 (b)
0.6 to 1.30 (b)
0.6 to 1.30 (b)
0.6 to 1.30 (b)
0.60 to 1.10 (b)
0.040
0.040
0.040
0.60 to 1.10 (b)

0.040
0.050
0.050
0.050
0.050
0.10 to 0.30
0.10 to 0.30
0.10 to 0.30
0.10 to 0.35
1 diameter
1 diameter
1 diameter
2 diameters
2 diameters
2 diameters
2 diameters
2 diameters
2 diameters
2 diameters
4 diameters
4 diameters
4 diameters
3 diameters
Silicon, per cent

4 diameters

(continued)
70

Item
ASTM A227-64
ASTM A227-68
ASTM A227-71
ASTM A648-72
"not applicable"
"not applicable"
alternative test (d)
alternative test (d)

Torsion test
Splitting Test
Item
ASTM A227-64
ASTM A227-68
ASTM A227-71
ASTM A648-72
200 min, 230 max.
200 min, 230 max.
200 min, 230 max.
200 min.
231 min, 261 max.
231 min, 261 max.
231 min, 261 max.
231 min.
262 min.

192 min, 221 max.
192 min, 221 max.
192 min, 221 max.
192 min.
222 min, 251 max.
222 min, 251 max.
222 min, 251 max.
222 min.
252 min.

Item
ASTM A648-84
ASTM A648-86a
ASTM A648-88a
ASTM A648-88b
1985
1987
1989
1990 (h)
0.45 to 0.75 (c)
0.45 to 0.75 (c)
Carbon, per cent

0.50 to 0.85 (c)
0.50 to 0.85 (c)
0.50 to 0.85 (c)
0.50 to 0.85 (c)
0.55 to 0.88 (c)
0.55 to 0.88 (c)
0.50 to 0.85 (c)
0.50 to 0.85 (c)
0.60 to 1.10 (b)
0.60 to 1.10 (b)
Manganese, percent
0.60 to 1.10 (b)
0.60 to 1.10 (b)
0.50 to 1.10 (b)
0.50 to 1.10 (b)
0.60 to 1.10 (b)
0.60 to 1.10 (b)
0.50 to 1.10 (b)
0.50 to 1.10 (b)
0.040
0.040
0.030
0.030
0.050
0.050
0.035
0.035
0.10 to 0.35
0.10 to 0.35
0.10 to 0.35
0.10 to 0.35
2 diameters
2 diameters
2 diameters
2 diameters
2 diameters
2 diameters
3 diameters
3 diameters
30% min.
30% min.
Silicon, per cent


71
(continued)

Item
ASTM A648-84
ASTM A648-86a
ASTM A648-88a
ASTM A648-88b
35% min.
35% min.
30% min.
30% min.
6/4/3 turns (e)
6/4/3 turns (e)
pass/fail
pass/fail

Torsion test
Splitting Test
pass/fail (g)

200 min, 230 max
200 min, 230 max
231 min, 261 max
231 min, 261 max
262 min, 297 max
262 min, 297 max
192 min, 222 max
192 min, 222 max

Item
ASTM A648-84
ASTM A648-86a
ASTM A648-88a
ASTM A648-88b
222 min, 252 max
222 min, 252 max
222 min, 252 max

(f)
222 min, 252 max (f)
252 min, 290 max
252 min, 290 max
252 min, 282 max

(f)
252 min, 282 max (f)
ASTM A648-90a
ASTM A648-94
ASTM A648-95
ASTM A648-04a
1994 (h)
1995 (j)
1998 (i) (k)
2004

Item
Carbon, per cent
0.50 to 0.85 (c)
0.50 to 0.85 (c)
0.50 to 0.85 (c)
0.50 to 0.85 (c)
0.50 to 0.85 (c)
0.50 to 0.85 (c)
0.50 to 0.85 (c)
0.50 to 0.85 (c)
Manganese, percent
0.50 to 1.10 (b)
0.50 to 1.10 (b)
0.50 to 1.10 (b)
0.50 to 1.10 (b)
0.50 to 1.10 (b)
0.50 to 1.10 (b)
0.50 to 1.10 (b)
0.50 to 1.10 (b)
0.030
0.030
0.030
0.030

Silicon, per cent
0.035
0.035
0.035
0.035
0.10 to 0.35
0.10 to 0.35
0.10 to 0.35
0.10 to 0.35
0.007 (m)

35% min.
35% min.
35% min.
35% min.
(continued)
72

30% min.
30% min.
30% min.
results reported
results reported
6/4/3 turns (e)
6/4/3 turns (e)
8/6/5 turns (e)(l)
10/8/7 turns (e) (l)
pass/fail
pass/fail
222 min, 252 max

(f)
222 min, 252 max

(f)
222 min, 252 max

(f)
222 min, 252 max (f)
252 min, 282 max

(f)
252 min, 282 max

(f)
252 min, 282 max

(f)
Item
ASTM A648-90a
ASTM A648-94
ASTM A648-95

Torsion test
Splitting Test
30% min.

252 min, 282 max (f)

ASTM A648-04a
Time to failure > 75
hrs. (g)
Notes
(a) Carbon in any one lot shall not vary more than 0.20 percent.
(b) Manganese in any one lot shall not vary more than 0.30 percent.
(c) Carbon in any one lot shall not vary more than 0.13 percent.
(d) " an alternative test procedure may be agreed upon "
(e) Varies by size (6, 1/4, 5/16) See Table 3. Value in parenthesis is equivalent tensile max on that basis.
(g) Supplementary requirement (S1.) applicable only when specified.
(h) This appears to include formatting and scope caveat revisions only.
(i) Reapproved 2000 and printed in 2001 standards.
(k) Added reference standards including AWWA C304.
Changing References to ASTM Wire Standards

AWWA Standard 7B.2-T (tentative) allowed wire manufactured to ASTM A229-41,
ASTM A227-47T, or ASTM A82-34. ASTM A82-34 (re-adopted through the 1952 edition) was
for Cold-Drawn Steel Wire for Concrete Reinforcement. The minimum tensile strength was
80,000 psi. The yield point of the wire was allowed to be established at 80 percent of the tensile
strength of the final rupture strength. ASTM A229-41 was for Oil Tempered Steel Spring
Wire. No yield strength was specified, rather a range of minimum and maximum tensile
strengths. ASTM A227-47T, for Hard Drawn Steel Spring Wire also specified only a range of
tensile strength. Comparing the last two standards, the minimum strength of ASTM A227-47 in
size 0.1192-inch wire was 192,000 psi, versus 190,000 psi for the oil tempered wire in ASTM
A229.
AWWA C301-52 allowed wire manufactured to ASTM A227 or ASTM A82.
AWWA C301-55 and AWWA C301-58 allowed wire manufactured to ASTM A227,
deleting ASTM A82 wire in the 1955 standard.
AWWA C301-64 and AWWA C301-72 allowed wire manufactured to ASTM A227 and
included except that wire with higher tensile strengths may be used if it meets the other
requirements of ASTM A227. In the ASTM A227-63T, effective at the time of adoption of
AWWA C301-64, there was only one range of tensile requirements (maximum to minimum.)
73
In ASTM A227-65, two strength classes were adopted. Those Class I and Class II tensile
requirements carried over to the ASTM A227-68, which was effective at the time of adoption of
AWWA C301-72.
AWWA C301-79 allowed wire manufactured to ASTM A648, a standard developed
exclusively for Steel Wire, Hard Drawing for Prestressing Concrete Pipe. The standard
included Wire with specified minimum tensile strengths exceeding those in A648, Class III may
be used if it meets the other requirements of ASTM A648, Class III. ASTM A648-72 includes
a footnote: Prior to the issuance of this specification it was common practice in the prestress
concrete pipe industry to refer to the requirements of Specification A 227, Hard Drawn Steel
Mechanical Spring Wire. ASTM A648-72 includes requirements nearly identical to ASTM
A227, except that a third tensile strength range, Class III was added.
AWWA C301-84 allowed wire manufactured to ASTM A648, but the exception allowing
higher strength wire was deleted. All subsequent standards incorporate ASTM A648. In ASTM
A648-88a, adopted in 1989, Class I wire, which was essentially the same as ASTM A227, was
dropped.
The 1992 standard revision addressed additional wire requirements above ASTM A648,
including a requirement for the wire manufacturer to audit the surface temperature of the wire
throughout the length of the wire-drawing process or take similar dependable precautions to
provide assurance that the maximum wire surface temperature does not exceed 360F (182C)
during drawing. Not specified was what constituted similar dependable precautions.
The minimum number of turns to failure in the torsion test was now specified to not be
less than 8 for 0.192-inch (4.88-mm) wire, 6 for 0.250-inch (6.35-mm) wire, and 5 for 0.312-inch
(7.92-mm) wire per 8 inches (203 mm) of test sample length. Torsion test specimens were
required to have cross section of the primary break showing approximately three-fourths or
more of the failure area as coplanar shear perpendicular to the wire axis. However, any coil for
which the failure torsion test sample has a radial, spiral (that is, longitudinal) split extending the
full length of the sample between the torsion machine jaws was allowed to be retested instead of
rejected.
In addition, relaxation-loss data were required to be provided as a means of qualifying
each manufacturer of wire to be used in the manufacture of pipe. Wire-relaxation-loss quality
assurance tests were also required in AWWA C304-92, Section 6.7.3.
Changes Within the ASTM Wire Standards
The differences between standard revisions were principally derived from the apparent
desire of the purchasers of that wire (the PCCP manufacturers) to minimize the amount of steel
necessary to prestress. Fundamentally, the higher the strength available from the wire, the less
wire needed to be included in the pipe and its economic advantage increased. How was the
increased strength achieved? It was achieved through advances in chemistry and mechanics.
Refer to Table 2.7. The upper limit on carbon was 0.75 percent up until 1964. ASTM
A227-64 included a new tensile strength range for each size of wire, and the new Class II wire
was allowed a range from 0.60 to 1.20 percent carbon. The upper limit on manganese was also
increased to 1.30 percent from 1.20 percent. ASTM A227-68 increased the upper limit on
carbon for the Class III, but decreased both the lower limit (to 0.45 percent) and the upper limit
74
(to 0.85 percent.) In ASTM A227-68, the carbon limit ranges for both classes of wire were from
0.45 to 0.85 percent. The major change in ASTM A227-71 was the increase in upper limit of
manganese for Class II to 1.30 percent.
With the development of ASTM A648-72, the new Class III wire was restricted to a
carbon range of between 0.55 and 0.93 percent. The lower bound of Class II carbon was
increased to 0.5 percent and the upper bound of Class I was reduced to 0.75 percent. With the
upper bound reduced slightly to 0.88 percent for Class III in ASTM A648-84 and further reduced
to 0.85 percent in ASTM A648-88a, the carbon limits have remained constant.
ASTM A648-84 contained significant changes to the standard. Including the carbon
limit revision and deletion of Class I, testing requirements were changed, and the minimum
strengths were expressed as minimum breaking strength in pounds, instead of calculated stress.
Also, the maximum amount of sulfur was reduced from 0.040 to 0.030 percent, and the
maximum silicon was reduced from 0.050 to 0.035 percent. Those limits are substantially
unchanged to date. Among the most important testing changes were the wire splitting test and
torsion test.
Commercial Changes Beyond the ASTM Wire Standards
It has been previously noted that AWWA C301-64 and AWWA C301-72 allowed wire
manufactured to ASTM A227 and included except that wire with higher tensile strengths
may be used if it meets the other requirements of ASTM A227. And when ASTM A648-72
was adopted by AWWA C301-79, the standard included Wire with specified minimum tensile
strengths exceeding those in A648, Class III may be used if it meets the other requirements of
ASTM A648, Class III. The net effect of this was for the wire manufacturers to try and achieve
higher tensile strength wire.
In particular, wire was manufactured by Interpace Corporation in its Solon, Ohio plant
and marketed as Class IV wire. There were even projects engineered to use Class III 1/2 wire,
somehow expected to be between the limits of the ASTM standards and the marketed Class IV
wire. In order to achieve these high strength levels, the wire was drawn through dies at a great
rate of speed, creating temperatures in excess of 400F. That proved to be detrimental to the
wire, both serving to induce longitudinal cracks and to make the wire particularly susceptible to
hydrogen embrittlement.
75
Figure 2.2 Total production PCCP by manufacturer
76
Figure 2.3 Total production PCCP ECP compared to LCP
77
Figure 2.4 Total ECP production by manufacturer
78
Figure 2.5 Total LCP production by manufacturer
79
Figure 2.6 Minimum cylinder thickness PCCP
Figure 2.7 Minimum wire size PCCP
80
Figure 2.8 Mortar coating - PCCP
81
CHAPTER 3
PERFORMANCE
CAUSES AND MODES OF FAILURE
Definition of PCCP Failure
The definition of failure utilized by the USBR in its 1995 AwwaRF study was:
Failure was indicated as requiring action after installation to correct a pipe deficiency
repair, replacement, or both repair and replacement of the affected units and The term
failure is synonymous with repair and replacement rate. (AwwaRF 1995)
Similar to the human population, each individual pipe section has a birth (manufacturing)
to death cycle that is affected by its heredity (design, manufacturer, materials, etc.), birth defects
(construction, installation), and lifestyle (operations and maintenance, etc.). For the purposes of
this study, failure has been defined as the loss of use of a pipe section or reduction in confidence
in that pipe section to remain service, after discovery of a pipe section deficiency. This includes
repair, replacement, or reduction in operating pressure.
Common Causes of PCCP Failure
Causes of failure of PCCP are numerous. Citations in the literature include high chloride
environment (Villalobos 1998), the quality of the mortar including lack of complete envelopment
of the prestressing wires within the cement mortar coating (Price, Lewis, and Erlin 1998), poor
quality of the reinforcing wires (Walsh and Hodge 1998, Knowles 1990), corrosive soils
(Galleher and Stift 1998), inadequate thrust restraint (Ojdrovic et al. 2001), construction damage
(Parks, Drager, and Ojdrovic 2001), cracks in the cylinder welds (Price 1990), and delamination
of the coating (Price 1990). Details of the failure mechanism of EC-PCCP with broken
prestressing wires have been published (Zarghamee and Ojdrovic 2001). Those details may be of
use only in the courtroom, however, because once the pipe condition has approached the
imminent failure state, it is too late.
PCCP is a rather complex engineered product which requires a good deal of attention
during the design, manufacturing, inspection, operation, and maintenance in order to be
successful. The following lists some of the many means that PCCP has failed, grouped by the
five components of PCCP pipeline success:
1.
Design had to have been as thorough as possible within the context of the stateof-the-art at that time and either engineered by the owner or well reviewed by the
owner.
H2S generated in unlined force mains has resulted in sulfuric acid attack of
the interior of PCCP. In retrospect, either the choice of PCCP was poor or the
detailing of the pipeline to avoid air pockets was poor.
Inadequate joint restraint has resulted in PCCP failures, particularly at the
joints, because of pipe that moved, exposing the joint to the environment. Some
mechanically restrained joints physically needed to move to engage the restraint
83
2.
mechanism, leading to the mortared joints opening and exposing the steel joint
rings to corrosive soil. Once the steel joint started to corrode, it expanded and
cracked the coating, exposing the wire and cylinder to the same soil and
groundwater.
Cantilever (bending or broken back) failures have occurred at structures
where the pipe was encased throughout its circumference. The dissimilarity in
bedding stiffness has led to differential settlement and circumferential cracks
clear through the pipe section.
Hydrotest pressures have been specified in excess of the design pressure
and design surge pressure, resulting in coating cracks that will never
autogeneously heal.
Design of PCCP for placement in soil with chlorides >700 ppm or in soil
with measured resistivities <3,000 ohm-cm without additional measures taken
may have led to premature failures due to the loss of the protective environment
around the prestressing wires.
The impact load of the pipe under its own weight on a rigid surface
providing negligible bedding angle can result in high unit stresses and cracking
of the core and cement mortar coating. It is still not standard practice to analyze
PCCP for these loadings.
Manufacturing by a manufacturer who was committed to making each length of
pipe to the highest quality level attainable, consistent with the design was and still
is critical.
Low qualities of mortar such as low density, low thickness, and low
cement content have all resulted in wire corrosion.
Leaking at joints may in many instances be attributable to out-ofroundness of the mating joints. Not just a misfit for undersize but cracking of the
core at the joint to the cylinder when the spigot was later forced into the bell by
the contractor. The significant stiffness discontinuity between the prestressed core
and the unstressed spigot ring also resulted in many pipe being shipped with
circumferential cracks in the thin part of the core at the junction of the cylinder
and the spigot.
Alkali-silica reactivity and poor concrete strength have been cited as
causes for low quality cores. The variation in compressive strength of PCCP cores
vertically cast also may have affected the variation in prestress noted along the
axis of some PCCP.
Wire manufacturing problems have been extensively reported.
Overheating of the wire during drawing leads to dynamic strain aging. That yields
poor torsion ductility and susceptibility to hydrogen embrittlement. Reports of
poor quality of wire were not confined to Type IV.
Excessive core creep and shrinkage can be as a result of too high a watercement ratio, too many fines in the aggregate, and inadequate or improper curing
of the concrete core prior to prestressing. The resulting excessive wire relaxation
is nothing like the design assumptions.
Excess coating shrinkage, often due to the same causes as described
above, result in exposure to wire reinforcement and subsequent corrosion. The
carbonated surface of the EC-PCCP core has a decreased pH of about 8.3, much
84
3.
4.
lower than the usual pH 12 of the surrounding mortar. That carbonation of the
concrete core surface in contact with the wire increases crevice corrosion. Also,
an uneven (layered) application of mortar results in a flushing of the protective
alkalinity from the around the wire.
Defective cylinders have been reported. The welds to the joint rings and
the lap (or joggle) joints have been poorly made.
Rough handling of the pipes when the coating is still green in the
manufacturers yard and similarly rough handling in loading, transport, and
unloading or placing in the trench may contribute to cracking of the coating.
Inspection by both manufacturer and owner, in the plant and during construction
is important.
Loss of prestress due to core under design strength at time of prestressing
has been previously noted. Loss of prestress due to wire breaks and splicing
probably should have led to rejection also.
PCCP leaking at joints may possibly be due to unnoticed construction
damage, missing joint coating, cracks in joint welds, a looped gasket, or poor
joint fit-up, all of which point to poor or no field inspection.
Dented cylinders have been reported, likely during fabrication. Active
inspection by the owner would have resulted in rejection at the plant.
Typical design bedding angle was 90 degrees. In actual practice, the
shaping of bedding to achieve 90-degree bedding requires extra effort and is a
requirement often times ignored. Failure to construct per the design results in
excess moments and thrusts resulting in cracked lining and coating.
Pipe has been sent from the factory with no effective wire stress, as
evident by examination of pipe with no residual prestress in the core. Upon
removal from the ground, wire has re-coiled to the original spool diameter,
indicating no permanent set. Similarly, wire removed has had as many as three
splices on a single pipe, resulting in very low levels of prestress.
Construction by a contractor motivated to comply with the requirements of the
construction documents is always important.
Wrong pipe class, caused by pipe laid out of order, has resulted in pipe
underdesigned and understrength for the location placed. This is not a fault of the
product but is indicative that each PCCP is typically designed for its specific
location in a pipeline and unsuitable for placement elsewhere.
Settlement, in general and at structures, is often the result of poor
compaction of bedding and backfill. Bedding not corresponding to design
assumptions is not unique to PCCP, but because of its strength in the hoop
direction rather than longitudinal for beam action, the correct placement of
bedding is critical to its long-term performance.
Construction of a PCCP pipeline in a manner different from what was
assumed may contribute to the structural distress of the pipeline. For example, the
coating on the pipe may be cracked if either the trench width or height of cover
exceed design assumptions.
Accidental penetration by third parties subsequent to construction and
mechanical damage during construction have been documented.
85
5.
Operations and maintenance (O&M) should not place the pipeline at additional
risk. Rapid changes in pressure due to vagaries in operation (surge) overstresses
the pipe. Transient waves smaller than design surge pressure have failed PCCP
with broken wires.
Wire is especially susceptible to hydrogen embrittlement (e.g., Class IV).
This is caused by the application of excessive cathodic protection, usually well
intentioned but in retrospect inappropriately, has caused many wires to break in
PCCP.
Excess external loads greater than the design assumptions and applied
subsequent to construction, usually by others, have resulted in PCCP pipelines
operated outside the design envelope.
How PCCP Fails

The most common failure scenario of PCCP with broken wires is hypothesized to be the
result of redistribution of hoop stress in the zone of no or ineffective prestressing wires to the
steel cylinder. As the core is relieved from stress when wires adjacent to the broken area break
under the extra load, it begins to crack. Internal inspection can usually detect this progressing
from visible longitudinal cracks. Mortar coating, if not already cracked, cracks in tension as the
thin cylinder expands radially under hoop stress. The cylinder is exposed to water through the
cracks, and corrosion of the cylinder progresses at a rate which is different for each pipe.
In other words, there is a cascading effect: overpressure coating cracks wires
exposed to water wires corrode and break pressure is transferred to the cylinder core
cracks cylinder is exposed to water cylinder corrodes and fails.
Failure mode of PCCP is usually sudden. The shear-friction of the concrete core is
exceeded simultaneously as the corroded cylinder ruptures in tension. The failure pressure is
lower than the ultimate bursting strength of the cylinder. It is limited to several factors including
variations in the thickness of the cylinder, eccentricities in the cylinder welds, weld strength, and
bending stresses at the transition to the much-stiffer core with functional prestress. If the pipe has
longitudinal forces applied through joint restraint, these forces compound the problem.
Failure of prestressing wires, and especially significant numbers of prestressing wires,
should be recognized as fair warning that the pipeline is operating at a significantly reduced
margin for safety and that its useful remaining life is finite.
Hydrostatic test pressures to which PCCP pipelines are tested are usually much lower
margin above operating pressure than 150 percent. For example, the 1961 edition of AWWA
Manual M9 Concrete Pressure Pipe states, Test pressures are commonly specified as some
value slightly greater than the operating pressure, such as 120 percent of operating pressure
(p. 103). It also states, Requiring a test pressure considerably in excess of the operating pressure
serves only to increase costs since the pipe strength and size of thrust blocks or harnessed joints
must be significantly increased.
The thrust restraint systems were likely also designed not for the hydrotest but only for
the operating condition (e.g., class 475-10, etc.). The test pressures at 150 percent of design
pressure would be about 130 percent in excess of that which would decompress the core and
cylinder with all of the wires effective. The hydrostatic pressure at 150 percent of operating
would also have exceeded the tensile stress limit for cracking the coating at the bottom and side
of the pipe and for cracking at the top as well. The conclusion is that if a PCCP pipeline was
86
hydrostatically tested at 150 percent of design pressure then the pipeline coating has likely been
cracked since that day.
DESIGN LIFE
Definition of Expected Service Life for PCCP
All pipeline materials have a finite useful life. The life expectancy varies by many
factors. PCCP life expectancy varies from 50 years, to 100 years, to indefinite depending on
the perception of the pipeline owner.
DATA PARTITIONS
As previously discussed in this study, PCCP has had a long and diverse history with
many changes in standards and materials. To account for these changes and any impact they
might have had on failure rates, the failure data was analyzed in groups of years that shared
similar standards and materials. These groups are as follows:
Pre-1955 (presumably as manufactured to AWWA 7B.2-T [1949])

19922007 (presumably as manufactured to AWWA C301-92, although the Principal
Investigators are aware of at least one project specified to AWWA C301-84 in 1994)
It is meaningful to understand that these partitions represent eras of similarly

manufactured pipe. With standards and materials being equal within each group of years, it is
assumed that there was no significant variation of failure rates within each data partition.
Classes of Failure
Three categories of PCCP failure have been defined:
1. Catastrophic ruptures and leaks (Category 1)
2. Significant deterioration or structural weakness discerned by inspection (Category 2)
by
Visual, sounding, and accidental discovery
Electronic inspections
3. Loss of service (Category 3)
Time out of service
Full or partial replacement
87
FAILURE DATABASE
The failure database has a total of 592 independent entries representing a diverse
collection of Category 1, 2, and 3 failures across all data partitions and across 35 states and the
District of Columbia.
To be included in the database, every entry represented one pipeline that had recorded at
least one failure and a corresponding location. No other lacking information precluded the entry
from the database. The failure data cataloged three forms of failure (catastrophic ruptures, failure
discerned by inspection, and failure through loss of service) in order to discern trends. It is
appropriate to clarify how those classifications were established. For example, one utility
reported that they had 18 ruptures and an additional 800 pipes failed (or are in need of repair due
to risk) from an inspection program. In that instance, those were cataloged as 18 separate
Category 1 failures and 800 Category 2 failures (broken wires discovered by nondestructive
testing [NDT] inspections). Other PCCP users have had ruptures, adjacent pipe failed due to
inspections, and had to condemn and abandon the use of thousands of feet of pipe in their
systems.
These damaged or defective pipes discovered during inspection and pipes no longer
deemed serviceable must be included in the calculations to get an accurate failure rate for the
pipes. But because the condemned and abandoned pipes had not ruptured or leaked, they could
not be cataloged as Category 1 failures. They had not been individually examined or inspected,
so they could not be classified as Category 2 failures. Thus, the creation of a third category,
Category 3, was warranted to account for these failures. However, it also became evident that
Category 2 failures were often the proximate determination of Category 3 failures, so from a
statistical analysis, they were lumped together, yielding an increased sample population size of the
combined category.
Of the 592 independent entries, the database includes 435 Category 1 failures, or 61.1
percent of the entries. There were 35,662 Categories 2 and 3 failures, or 45.3 percent of the
entries. Thirty-eight of the entries, or 6.5 percent, specified both a Category 1 and a Categories 2
and 3 failure. Of all the entries, 98.3 percent of the entries had a diameter specified, 97.8 percent
had a pipe type (embedded, lined, etc.) specified or assumed based on the diameter, and 92.9
percent had a wire type (Class I, II, III, or IV) specified or assumed based on the installation
year. In totality, 95.9 percent had the installation date specified and 92.0 percent had a fail date
specified. With these, a total of 85.6 percent of the entries had both the installation date and the
fail date, such that a pipe age could be calculated.
Looking at Category 1 failures by manufacturer, 41.8 percent of the entries were
tabulated as Interpace Corporation pipe, 9.0 percent was attributed to other manufacturers by
name, and the remaining 49.2 percent was unknown. Of the Categories 2 and 3 failures, 60.7
percent of the entries were Interpace pipe, 2.7 percent was attributed to other manufacturers, and
the remaining 36.6 percent had no manufacturer recorded.
Breaking the database into the partitions in a meaningful manner required the pipeline
age. This left the sample populations shown in Table 3.1 for further analysis:
88
Table 3.1
Sample populations
Partition
Category 1 Failures
All samples
Categories 2 and 3 Failures
393
24,822
pre-1955
32
10
195563
40
2,381
196467
31
63
196871
60
46
194
15,158
152
4,349
14
2,586
35
5,864
1,299
197278 (all)
197278 (Interpace)
197278 (non-Interpace)
197991
19922007
It should be noted that the subset 1972-78 (Interpace) included all failures of pipe known
to be manufactured by Interpace Corporation. However, the subset 1972-78 (unknown and nonInterpace) included both known non-Interpace manufactured pipe and unknown pipe; unknown
pipe may be either Interpace or non-Interpace manufactured pipe.
HISTOGRAMS AND SIMPLE STATISTICAL ANALYSIS
Using the above-listed sample populations of entries with pipeline ages, lifespan
histograms were produced. A histogram is a convenient graphical representation of a frequency
distribution. In this case, the frequency distribution of failures was plotted as a function of the
pipeline age for each of the corresponding data partitions.
PCCP Failures by Age
The first histogram, Lifespan Histogram-All Samples, Figure 3.1, illustrates the
breakdown of failures into five-year incremented age categories. This histogram disregards
which data partition to which each sample belongs, but is useful to observe the overall trend of
all failures. On the left-hand axis, the solid blue bars represent Category 1 failures, or ruptures.
The right-hand axis corresponds with the crossed magenta bars, or Categories 2 and 3 failures.
The mode, or value that occurs most frequently, of Category 1 failures occurs in 6 to 10 years,
while Categories 2 and 3 failures peak at the 26- to 30-year group. Both datasets are roughly
normally distributed, or bell-shaped, with a moderate skew to the right for the Category 1
failures and a slight skew to the left for Categories 2 and 3 failures. The Categories 2 and 3
failure distribution generally lags the Category 1 failure distribution by 20 years. Although not
perfectly normally distributed, the mean and standard deviation of both datasets were calculated.
The means were 13.95 and 16.75 years for Category 1 and Categories 2 and 3 failures,
respectively. The corresponding standard deviations were 8.95 and 10.24 years. By examining
the total production of pipes, a failure rate for each dataset was calculated. With a total of
4,979,837 pipe (or individual pipe segments) produced between 1940 and 2006, the average
failure rates for Category 1 and Categories 2 and 3 were -7.89 x 10-5 and 4.98 x 10-3 failures per
pipe produced, respectively. This indicates that within 50 years of being installed, one rupture
89
and 66 Category 2 or Category 3 failures occurred for every 13,200 pipe segments (~50 miles of
pipe).
The next histogram, Figure 3.2, examines those failed pipelines that were installed before
1955. For this era, the wire used was most probably Class I, 0.125-inch size, 45 percent design
stress. Again, age is represented in five-year increments on the bottom abscissa. The
approximate corresponding year is labeled below the age increments. The left-hand axis
corresponds with the solid blue bars which are Category 1 failures. The right-hand axis
corresponds with the crossed magenta bars, or Categories 2 and 3 failures. It is observed that the
mode of both Category 1 and Categories 2 and 3 failures occurs in 21 to 25 years. Both datasets
appear roughly normally distributed until after 31 to 35 years, with no failures occurring beyond
this timeframe. The mean number of years for failures was 23.32 and 24.86, respectively, for
Category 1 and Categories 2 and 3 failures. The corresponding standard deviations were 6.92
and 5.96 years. With 476,458 pipe produced in this timeframe, the failure rates were 6.72 x 10-05
and 2.10 x 10-05 Category 1 and Categories 2 and 3 failures per pipe segment produced,
respectively. This indicates statistically that within 35 years of being installed, one rupture and
315 other failures occurred for every 15,000 pipe segments (~57 miles of pipe). After 35 years of
age, there are no reported failures.
The histogram following pre-1955, Figure 3.3, is for pipelines installed between 1955
and 1963. For this timeframe, the wire used was most probably Class I, 6 gauge, stressed to 45 to
70 percent of the ultimate strength. The axes are the same as above. In this histogram, a bimodal
distribution exists for both Category 1 and Categories 2 and 3 failures, occurring at 25 to 30 and
41 to 45 years. Although less meaningful in this type of distribution, the mean and standard
deviation for Category 1 is 23.78 years and 10.21 years. For Categories 2 and 3, the mean is
29.79 and the standard deviation is 10.26. With 1,051,498 pipes produced between 1955 and
1963, the corresponding failure rates are 3.8 x 10-5 and 2.26 x 10-3 failures per pipe produced for
Category 1 and Categories 2 and 3 failures, respectively. This means for every 35,800 pipe
segment (~135 miles of pipe), there was one rupture and 81 Category 2 and Category 3 failures.
Figure 3.4, Lifespan HistogramInstalled 196467, looks at pipelines installed in this era.
This timeframe represents the first time Class II wire is listed in the standard. Class II, 8 gauge,
stressed to 75 percent ultimate tensile strength is the probable wire type used during this era. Age
is again represented in five-year increments on the bottom abscissa. This is the first graph where
present time is being approached, illustrated as being grayed out on the right-hand side. The
approximate corresponding year is labeled below the age increments. The left-hand axis
corresponds with the solid blue bars which are Category 1 failures. The right-hand axis
corresponds with the crossed magenta bars which are Categories 2 and 3 failures. Category 1
failures are roughly normally distributed around the mode of 11 to 15 years, 10 years earlier than
pipes produced during the previous era with Class I wire. Categories 2 and 3 failures are closer
to being uniform than normally distributed, unless 0 to 5 and >30 years are neglected. The means
were 15.13 and 16.90 years and the standard deviations were 6.71 and 10.06 years for Category
1 and Categories 2 and 3, respectively. With 594,367 pipes being produced in this timeframe, the
failure rates were 5.22 x 10-05 and 1.06 x 10-04, respectively, for Category 1 and Categories 2 and
3 failures. This equates to one rupture and two Category 2 or Category 3 failures for every
20,000 pipe segments (~75 miles of pipe).
90
Figure 3.5 illustrates failures of pipe installed in 196871. This timeframe also includes the
first time a new wire class is listed in the standard, Class III. The most probable wire used in this
era was Class III, 8 gauge, stressed to 75 percent ultimate strength. Age is still represented in fiveyear increments on the bottom x-axis. Present time is sill encroaching on the right-hand side of the
graph; future time is illustrated as being grayed out. The approximate corresponding year is labeled
below the age increments. Again, the left-hand axis corresponds with the solid blue bars which
represent Category 1 failures. The right-hand axis corresponds with the crossed magenta bars
which represent Categories 2 and 3 failures. Category 1 failures are more or less normally
distributed around the mode of 6 to 10 years, 5 years earlier than pipes produced during the
previous era with Class II wire. Again, Categories 2 and 3 failures are closer to being uniform than
normally distributed. The mean number of years for failures was 11.07 and 13.47, respectively, for
Category 1 and Categories 2 and 3 failures. The corresponding standard deviations were 6.01 and
8.14 years. With 551,345 pipes produced between 1968 and 1971, the failure rates were 1.09 x 1004
and 8.34 x 10-05 Category 1 and Categories 2 and 3 failures per pipe produced, respectively. This
indicates that within 40 years of being installed, one rupture and about one Category 2 or Category
3 failure occurred for every 9,200 pipes (~35 miles of pipe).
The next three graphs represent failures occurring in pipe installed in 197278. This
timeframe represents an era when a loophole was used by Interpace Corporation and others to
use higher strength wire, Class IV, while still adhering to the standard. Figure 3.6 looks at all
pipe failures from this era, Figure 3.7 only looks at Interpace pipe, and Figure 3.8 looks at both
Interpace pipe and pipe with unknown pipe manufacturer. It is noteworthy that over three times
as many Category 1 failures occurred from this time period than from the last timeframe.
Seventy-eight percent of the failures represented Interpace pipe and the remaining twenty-two
percent represented non-Interpace pipe and pipe of unknown manufacturer. The most probable
wire used in this era was Class III for non-Interpace pipe or Class IV for Interpace pipe, 8 gauge,
75 percent ultimate strength. For Category 1 failures, the all-samples (Figure 3.6), Interpace
(Figure 3.7), and unknown/non-Interpace (Figure 3.8) graphs are all roughly normally
distributed, with moderate skew to the right. The all-samples and Interpace graphs are modal
around the 6- to 10-year point, markedly similarly to the last timeframes Class III wire. The
unknown/non-Interpace graph is roughly modal around the 16-20 point. For the all-samples
graph, the Categories 2 and 3 failures are roughly normally distributed around the 26- to 30-year
point with a slight skew to the left. A similar distribution is observed in the Interpace pipe graph,
except being modal around 21 to 25 years. The unknown/non-Interpace graph has a modal
distribution of Categories 2 and 3 failures at 26 to 30 years. The Category 1 means were 12.16,
10.38, and 17.90 years for all-samples, Interpace, and unknown/non-Interpace, respectively. The
corresponding standard deviations were 7.70, 6.47, and 8.56 years. The Categories 2 and 3
means were 15.97, 12.44, and 22.62 years, again for all-samples, Interpace, and unknown/nonInterpace, respectively. The corresponding standard deviations were 9.75, 8.45, and 8.56 years.
The production of PCCP approached 856,323 pipes overall for this timeframe, with
468,296 coming from Interpace. This corresponds with Category 1 failure rates of 2.27 x 10-04,
3.25 x 10-04, and 4.90 x 10-05 failures per pipe produced for all-samples, Interpace, and nonInterpace, respectively. The difference between Interpace and unknown/non-Interpace failure
rates is over an order of magnitude larger. The corresponding Categories 2 and 3 failures for the
years 197278 were 1.77 x 10-02, 9.29 x 10-03, and 1.26 x 10-02 failures per pipe produced for allsamples, Interpace, and unknown/non-Interpace. Statistically, for 22,000 pipes (~83 miles)
installed in this era, one rupture occurred if the pipe was unknown/non-Interpace, seven ruptures
91
occurred if the pipe was Interpace, and five ruptures occurred among all pipe manufactured in
the era. For Categories 2 and 3 failures, for the same 22,000 pipes installed, 277 other failures
occurred if the pipe was unknown or non-Interpace pipe, 204 failures occurred if the pipe was
Interpace, and 389 failures occurred among all pipe manufactured in the era.
Figure 3.9, Lifespan HistogramInstalled 197991, represents the end of the Class IV
loophole. Class III, 8 gauge, stressed to 75 percent ultimate tensile strength is the probable wire
type used during this era. Age is again represented in five-year increments on the bottom
abscissa. Almost half of the graph is future time, illustrated by being grayed out on the righthand side. The approximate corresponding year is labeled below the age increments. The lefthand axis corresponds with the solid blue bars which signify Category 1 failures. The right-hand
axis corresponds with the crossed magenta bars which signify Categories 2 and 3 failures.
Category 1 failures are J-curved to the right, with the highest frequency of failures occurring in 0
to 5 yearsfive years earlier than the Class III/IV wire used in the previous timeframe.
Categories 2 and 3 failures are approximately normally distributed around 21 to 25 years,
similarly to the Categories 2 and 3 failures of Interpace pipe during the last era. The mean
number of years for failures was 7.47 and 12.72, respectively, for Category 1 and Categories 2
and 3 failures. The corresponding standard deviations were 6.37 and 7.77 years. With 1,067,552
pipes produced in this timeframe, the failure rates were 3.28 x 10-05 and 5.49 x 10-03 Category 1
and Categories 2 and 3 failures per pipe produced, respectively. This indicates that within 30
years of being installed, one rupture and 176 Category 2 or Category 3 failures occurred for
every 32,000 pipes (~120 miles of pipe).
The last lifespan histogram, Figure 3.10, representing the timeframe 19922007, is
included for completeness. With one Category 1 failure and 1,299 Categories 2 and 3 failures
from one database entry, the sample population is not statistically significant. Class III wire,
0.192 inch, stressed to 75 percent ultimate strength is the most probable wire installed in this
timeframe. Seventy percent of the graph represents future time, illustrated as being grayed out on
the right-hand side of the graph. With only a single entry (one pipeline that failed and was
subsequently relined), the failure peak is depicted at the 11- to 15-year point for both Category 1
and Categories 2 and 3 failures. The mean is 12 years, with no standard deviation. With
382,295 pipes produced between 19922006, the corresponding failure rate was 2.62 x 10-06 and
3.40 x 10-03 Category 1 and Categories 2 and 3 failures per pipe produced, respectively.
PCCP Failures by Installation Date
The next four histograms, Figures 3.11 through 3.14, illustrate Category 1 and
Categories 2 and 3 failures between 1942 and 2006 by installation date instead of by pipeline
age. Because only the installation year and failure data are required to construct these
histograms, several more data points are available. The sample population was represented by
403 Category 1 failures and 27,805 Categories 2 and 3 failures.
The first two histograms (Figures 3.11 and 3.12) represent Category 1 failuresthe first
without superimposed corresponding production, the second with. The next two histograms
(Figures 3.13 and 3.14) represent Categories 2 and 3 failuresagain, the third without
superimposed production, the fourth with the production values. Installation year is incremented
on the bottom abscissa throughout. On Figures 3.11 and 3.12, the left-hand axis corresponds with
the solid blue bars which signify Category 1 failures. On Figures 3.13 and 3.14, the left-hand
axis corresponds with the solid magenta bars which signify Categories 2 and 3 failures.
92
Additionally, on Figures 3.13 and 3.14, production data in individual pipe segments (sticks) has
been superimposed, corresponding to the horizontally lined, coral-colored right-hand axis.
On Figures 3.11 and 3.12, the Category 1 failures are normally distributed around the
year 1974; that is, the highest frequency of failures occurred from pipe installed in 1974. There is
a slight skew to the left. On Figures 3.13 and 3.14, the frequency is roughly normally distributed
around the installation year 1976. The mean number of failures between 1942 and 2006 were
6.20 and 428 Category 1 and Categories 2 and 3 failures, respectively. The corresponding
standard deviations were 10.23 and 1,194 failures. Although the production values are
superimposed on the second and fourth graph, they are not taken into account when calculating
the failure frequency distribution. Since failures are inherently a function of how much pipe was
produced, looking at the failures alone may lead to erroneous conclusions.
The next two graphs attempt to rectify this issue. Figures 3.15 and 3.16, named
Normalized Failures by Installation Date Histogram, Category 1 and Categories 2 and 3, the
failure frequency has been divided by the production in pipe for each corresponding installation
year. The result is the normalized frequency of failures per pipe produced on the left-hand
ordinate, plotted as a function of installation year, which runs across the bottom abscissa. The
solid blue bars signify normalized Category 1 failures; the crossed magenta bars signify
normalized Categories 2 and 3 failures. By taking into account production, the Category 1 failure
distribution became bimodal: one peak occurred in 1975, the other in 1947. For Categories 2 and
3 normalized failures, the distribution is roughly normally distributed between 1975 and 1979.
The fact that a mode still occurs in the mid-1970s for both Category 1 and Categories 2 and 3
confirmed the validity of the previous histograms by installation year that did not take into
account production. The mean was 6.09 x 10-05 Category 1 failures per pipe produced with a
corresponding standard deviation of 9.17 x 10-05 failures per pipe produced. For Categories 2 and
3, the mean was 4.01 x 10-03 failures per pipe produced with a corresponding standard deviation
of 1.12 x 10-02 failures per pipe produced.
PCCP Failure Rates by Manufacturing Era
Figure 3.17 summarizes the PCCP failure rates as a function of timeframe, or era of pipe
manufacturing. Figure 3.17 is plotted with standard axes. Like the preceding histograms, on
Figure 3.17 the left-hand axis of the bar graph corresponds with the solid blue bars which
represent Category 1 failures. The right-hand axis corresponds with the crossed magenta bars
which represent Categories 2 and 3 failures. The pipe installation timeframe increases along the
bottom x-axis. Observing the failure rate trends in Figure 3.17, it can be seen that the largest
Category 1 failure rate occurred in the 197278 era with Interpace pipe. The largest Categories 2
and 3 failure rate occurred in the 197278 era with all samples. The most agreement between
Category 1 and Categories 2 and 3 failures occurred in 196467 and 1968-71 timeframes.
PCCP Failures by Wire Class
Figure 3.18 illustrates PCCP failures as a function of wire class. Because pipeline age
and installation year are not required for this analysis, several more database entries are
available. The sample populations for this histogram are 397 and 25,809 Category 1 and
Categories 2 and 3 failures, respectively. The left-hand axis corresponds with the solid blue bars
which represent Category 1 failures. The right-hand axis corresponds with the crossed magenta
bars which represent Categories 2 and 3 failures. Wire class increases across the x-axis. It is
93
observed that the highest frequency of Category 1 failures occurred with Class IV wire. The
second highest frequency of failures occurred with Class II wire. For Categories 2 and 3 failures,
the highest occurrence of failures also occurred with Class IV wire. The second highest
frequency of failures occurred with Class III wire. The least number of Categories 2 and 3
failures were reported for Class II wire.
PCCP Failure Rates by Pipe Type by Installation Year
Figures 3.19 and 3.20, Failure Rates by Pipe Type by Installation Year for Category 1
failures and Categories 2 and 3 failures, examine failure rates between ECP and LCP as a
function of the installation year. Because another parameter was added for this analysis, a
decrease in the sample population resulted. The sample population was 387 and 24,400 for
Category 1 and Categories 2 and 3, respectively.
Figure 3.19 observes Category 1 failures. The y-axis represents Category 1 failure rates,
or the number of failures divided by the pipe produced per installation year. The x-axis
represents the pipe installation year. The solid blue bars represent LCP failure rates. The
horizontally-lined coral-colored bars represent ECP failure rates.
In Figure 3.20, the y-axis represents Categories 2 and 3 failures rates. For Category 1
failures, it is observed that LCP has a bimodal distributionone peak in 1947, the other in 1975.
The 1975 peak is roughly normally distributed; the 1947 peak is J-curved to the right. ECP is
roughly normally distributed about the mean occurring in 1972.
The average failure rates between 1942 and 2006 were 0.0126 and 0.0188 failures per
mile per year for LCP and ECP, respectively. Figure 3.20, representing Categories 2 and 3
failure rates, illustrates the trends for ECP and LCP. There were over five times as many
reported ECP failures as LCP failures, although more than twice as much LCP was installed
between 1942 and 2006. No orderly distribution was discernable for LCP. For ECP, the failure
rates were more or less normally distributed around 1975. The average failure rates between
1942 and 2006 were 0.254 and 2.65 failures per mile per year for LCP and ECP, respectively.
94
AwwaRF 4034 - Failure of PCCP

Lifespan Histogram
All Samples
12,000
140
Category 1
Sample pop. = 393 failures
Total pop. = 4,979,837 sticks
Failure rate = 7.89E-05 failures/
sticks produced
Mean = 13.95 years
= 8.95 years
100
10,000
Category 2 & 3
Sample pop. = 24,822 failures
sticks produced
Mean = 16.75 years
= 10.24 years
80
8,000
6,000
60
4,000
40
2,000
20
0
0-5
6-10
11-15
16-20
21-25
26-30
Age (years)
Figure 3.1 Lifespan histogram all samples
31-35
36-40
41-45
46-50
Frequency (no. of category 2 & 3 failures)
95
Frequency (no. of category 1 failures)
120

Lifespan Histogram
Installed pre-1955
Wire = Class I, 0.125", 45% ULT
12
Category 1
Total pop. = 476,458 sticks
sticks produced
Mean = 23.32 years
= 6.92 years
Category 2 & 3
sticks produced
Mean = 24.86 years
= 5.96 years
6
3
4
2
0
0-5
6-10
11-15
16-20
21-25
26-30
31-35
36-40
41-45
~1983
~1988
46-50
Age (years)
~1948
~1953
~1958
~1963
~1968
~1973
~1978
Approximate corresponding date
Figure 3.2 Lifespan histogram installed pre-1955
~1993
96
10

Lifespan Histogram
Installed 19551963
Wire = Class I, 6 ga, 4570% ULT
1,400
12
97
1,200
Category 2 & 3
sticks produced
Mean = 29.79 years
= 10.26 years
1,000
800
6
600
4
400
200
0
0-5
6-10
11-15
16-20
21-25
26-30
31-35
36-40
41-45
46-50
Age (years)
~1959
~1964
~1969
~1974
~1979
~1984
~1989
Figure 3.3 Lifespan histogram installed 1955-1963
~1994
~1999
~2004
Category 1
sticks produced
Mean = 23.78 years
= 10.21 years
10

Lifespan Histogram
Installed 19641967
Wire = Class II, 8 ga, 75% ULT
16
Category 1
sticks produced
Mean = 15.13 years
= 6.71 years
20
15
14
12
10
8
10
Category 2 & 3
sticks produced
Mean = 16.90 years
= 10.06 years
0
0-5
6-10
11-15
16-20
21-25
26-30
31-35
36-40
41-45
46-50
Age (years)
~1966
~1971
~1976
~1981
~1986
~1991
~1996
~2001
~2006
~2011
98
25

Lifespan Histogram
Installed 19681971
Wire = Class III, 8 ga, 75% ULT
40
16
Category 1
sticks produced
Mean = 11.07 years
= 6.01 years
30
25
20
14
12
10
Category 2 & 3
sticks produced
Mean = 13.47 years
= 8.14 years
15
10
0
0-5
6-10
11-15
16-20
21-25
26-30
31-35
36-40
41-45
46-50
Age (years)
~1970
~1975
~1980
~1985
~1990
~1995
~2000
~2005
~2010
~2015
99
35

Lifespan Histogram
Installed 19721978
Wire = Class III/IV, 8 ga, 75% ULT
70
8,000
100
50
7,000
Category 2 & 3
sticks produced
Mean = 15.97 years
= 9.75 years
6,000
5,000
40
4,000
30
3,000
20
2,000
10
1,000
0
0-5
6-10
11-15
16-20
21-25
26-30
31-35
36-40
41-45
~2010
~2015
46-50
Age (years)
~1975
~1980
~1985
~1990
~1995
~2000
~2005
~2020
Category 1
sticks produced
Mean = 12.16 years
= 7.70 years
60

Lifespan Histogram
Installed 19721978: Interpace
Wire = Class IV, 8 ga, 75% ULT
Category 1
sticks produced
Mean = 10.38 years
= 6.47 years
101
50
40
2,000
Category 2 & 3
sticks produced
Mean = 12.44 years
= 8.45 years
1,500
30
1,000
20
500
10
0
0-5
6-10
11-15
16-20
21-25
26-30
31-35
36-40
41-45
46-50
Age (years)
~1975
~1980
~1985
~1990
~1995
~2000
~2005
Figure 3.7 Lifespan histogram installed 1972-1978: Interpace
~2010
~2015
~2020
2,500
60

Lifespan Histogram
Installed 19721978: unknown & non-Interpace
Wire = Class III or IV, 8 ga, 75% ULT
14
7,000
Category 1
sticks produced
Mean = 17.90 years
= 8.56 years
10
6,000
5,000
4,000
Category 2 & 3
sticks produced
Mean = 22.62 years
= 8.56 years
3,000
2,000
1,000
0
0-5
6-10
11-15
16-20
21-25
26-30
31-35
36-40
41-45
46-50
Age (years)
~1975
~1980
~1985
~1990
~1995
~2000
~2005
Figure 3.8 Lifespan histogram installed 1972-1978: non-Interpace
~2010
~2015
~2020
102
12

Lifespan Histogram
Installed 19791991
Wire = Class III, 8 ga, 75% ULT
16
3,000
Category 1
sticks produced
Mean = 7.47 years
= 6.37 years
12
10
2,500
Category 2 & 3
sticks produced
Mean = 12.72 years
= 7.77 years
2,000
1,500
6
1,000
4
500
2
0-5
6-10
11-15
16-20
21-25
26-30
31-35
36-40
41-45
46-50
Age (years)
~1985
~1990
~1995
~2000
~2005
~2010
~2015
~2020
~2025
~2030
103
14

Lifespan Histogram
Installed 19922007
Wire = Class III, 0.192", 75% ULT
1.2
1,400
Category 1
Sample pop. = 1 failure
sticks produced
Mean = 12 years
= N/A
0.8
1,200
1,000
Category 2 & 3
sticks produced
Mean = 12 years
= N/A
0.6
800
600
0.4
400
0.2
200
0
0-5
~2000
6-10
~2005
11-15
~2010
16-20
~2015
21-25
26-30
31-35
Age (years)
~2020
~2025
~2030
36-40
~2035
41-45
~2040
46-50
~2045
104

Failures by Installation Date Histogram
Category 1, 19422006
50
45
Category 1
Mean = 6.20 failures
= 10.23 failures
35
30
25
20
15
10
Installation year
Figure 3.11 Failures by installation date histogram Category 1, 1942-2006
2006
2004
2002
2000
1998
1996
1994
1992
1990
1988
1986
1984
1982
1980
1978
1976
1974
1972
1970
1968
1966
1964
1962
1960
1958
1956
1954
1952
1950
1948
1946
1944
0
1942
105
40

Failures by Installation Date with Production Histogram
50
200,000
45
180,000
160,000
Category 1
= 10.23 failures
140,000
Installation year
Figure 3.12 Failures by installation date with production histogram Category 1, 1942-2006
2006
2004
2002
2000
1998
1996
1994
1992
1990
1988
1986
1984
1982
1980
1978
1976
1974
1972
1970
0
1968
0
1966
20,000
1964
1962
40,000
1960
10
1958
60,000
1956
15
1954
80,000
1952
20
1950
100,000
1948
25
1946
120,000
1944
30
Production (sticks)
35
1942
106
40

Failures by Installation Date Histogram
Category 2 & 3, 19422006
6,000
Category 2 & 3
= 1193.92 failures
4,000
3,000
2,000
1,000
Installation year
Figure 3.13 Failures by installation date histogram Categories 2 and 3, 1942-2006
2006
2004
2002
2000
1998
1996
1994
1992
1990
1988
1986
1984
1982
1980
1978
1976
1974
1972
1970
1968
1966
1964
1962
1960
1958
1956
1954
1952
1950
1948
1946
1944
0
1942
107
5,000

Failures by Installation Date with Production Histogram
Category 2 & 3, 19422006
200,000
6,000
180,000
5,000
120,000
100,000
3,000
80,000
2,000
60,000
40,000
1,000
20,000
Installation year
Figure 3.14 Failures by installation date with production histogram Categories 2 and 3, 1942-2006
2006
2004
2002
2000
1998
1996
1994
1992
1990
1988
1986
1984
1982
1980
1978
1976
1974
1972
1970
1968
1966
1964
1962
1960
1958
1956
1954
1952
1950
1948
1946
1944
Production (sticks)
140,000
Category 2 & 3
= 1193.92 failures
4,000
1942
108
160,000

Normalized Failures by Installation Date Histogram
4.0E-04
3.5E-04
Category 1
Mean = 6.09E-05 failures / sticks produced
= 9.17E-05 failures / sticks produced
3.0E-04
2.5E-04
2.0E-04
1.5E-04
1.0E-04
5.0E-05
Installation year
Figure 3.15 Normalized failures by installation date histogram Category 1, 1942-2006
2006
2004
2002
2000
1998
1996
1994
1992
1990
1988
1986
1984
1982
1980
1978
1976
1974
1972
1970
1968
1966
1964
1962
1960
1958
1956
1954
1952
1950
1948
1946
1944
0.0E+00
1942
109
Normalized frequency (no. of failures per sticks produced)
4.5E-04

Normalized Failures by Installation Date Histogram
Category 2 & 3, 19422006
0.06
Category 2 & 3
Mean = 4.01E-03 failures / sticks produced
= 1.12E-02 failures / sticks produced
0.05
0.04
0.03
0.02
0.01
Installation year
Figure 3.16 Normalized failures by installation date histogram Categories 2 and 3, 1942-2006
2006
2004
2002
2000
1998
1996
1994
1992
1990
1988
1986
1984
1982
1980
1978
1976
1974
1972
1970
1968
1966
1964
1962
1960
1958
1956
1954
1952
1950
1948
1946
1944
0
1942
110
Normalized frequency (no. of failures per sticks produced)
0.07

Failure Rates by Time Period
3.50E-04
2.00E-02
1.60E-02
2.50E-04
1.40E-02
1.20E-02
2.00E-04
1.00E-02
1.50E-04
8.00E-03
6.00E-03
1.00E-04
4.00E-03
5.00E-05
2.00E-03
0.00E+00
0.00E+00
1942-2007 pre-1955 195563
196467
All
196871
197278
All
197278
Interpace Unknown
& nonInterpace
Time period of production
Figure 3.17 Failure rates by time period
197278
197991 19922007
Failure rate (no. of category 2 & 3 failures per

sticks produced)
111
Failure rate (no. of category 1 failures per sticks

produced)
1.80E-02
3.00E-04

Failures by Wire Class Histogram
10,000
200
180
9,000
Category 1
112
Category 2 & 3
140
7,000
120
6,000
100
5,000
80
4,000
60
3,000
40
2,000
20
1,000
0
0
Class I
Class II
Class III
Wire Class
Figure 3.18 Failure by wire class histogram
Class IV
8,000
160

Failure Rates by Pipe Type By Year Installed Histogram
0.16
LCP
Total pop. = 13,458 miles
Mean failure rate = 0.0126
failures per mile per year
0.14
0.12
ECP
0.1
0.08
0.06
0.04
0.02
Installation year
Figure 3.19 Failure rates by pipe type by year installed histogram Category 1, 1942-2006
2006
2004
2002
2000
1998
1996
1994
1992
1990
1988
1986
1984
1982
1980
1978
1976
1974
1972
1970
1968
1966
1964
1962
1960
1958
1956
1954
1952
1950
1948
1946
1944
0
1942
113
Failure rate (no. of category 1 failures per miles produced)
0.18

Failure Rates by Pipe Type By Year Installed Histogram
Category 2 & 3, 19422006
40
LCP
35
30
ECP
25
20
15
10
Installation year
Figure 3.20 Failure rates by pipe type by year installed histogram Categories 2 and 3, 1942-2006
2006
2004
2002
2000
1998
1996
1994
1992
1990
1988
1986
1984
1982
1980
1978
1976
1974
1972
1970
1968
1966
1964
1962
1960
1958
1956
1954
1952
1950
1948
1946
1944
0
1942
114
Failure rate (no. of category 2 & 3 failures per miles produced)
45
CHAPTER 4
ASSESSMENT MODEL
ASSESSMENT METHODS TYPES AND RESULTS
Risk is the combined measure of the chance of a particular asset failing (condition based)
and the result of that failure. Calculation of risk for any event is based on a very simple formula:
Risk = Likelihood (Probability of Occurrence) Consequence
It is important to note that risk is related to not only condition of an asset but the impact
of that event. The project at hand can only address the likelihood portion of the risk equation.
The consequence of a failure is unique to a utility and pipeline.
Classically, risk can be quantitatively or qualitatively determined. Qualitative risk
assessments require relative information and agreement on the likelihood (probability) and
consequence of an event. Quantitative risk assessments require detailed information generally
not available until risk management programs have been initiated using qualitative rankings. For
the purposes of this study, only qualitative risk assessments of corrosion event have been
considered. The creation of a risk assessment model is intended to increase operational integrity
of the system and the credibility of the operator, especially when such techniques are offered for
public viewing. In a concerted effort to demonstrate its dedication to public and environmental
safety, while simultaneously seeking to reduce potential liabilities and regulatory requirements,
the utility should investigate a number of risk management alternatives taking what measures
seem appropriate for the complex and unique system it operates.
WORKSHOP WITH UTILITIES
The workshop was expected to accomplish the following:
Introduce project participants to facilitate future discussions and participation in the

project. (Relationships developed during the workshop will be extremely important,
since participants will be asked to review outlines, draft products, act as general
consultants to the work, and participate in testing of the concepts.)
Refine and add to risk defining elements of PCCP design, manufacturing, installation,
inspection, and operational parameters.
Refine general risk weighting concepts for PCCP.
Document the hurdles, and possibly the fatal flaws, that exist to development of each
concept.
Generate interest and enthusiasm for subsequent development of any tools. (Note that
full development of a tool or technique may involve the investment of several million
dollars, in which case an entrepreneur may ultimately need to step forward following
conclusion of the AwwaRF project.)
115
The full-day workshop was held with participating utilities on January 31, 2007 at the
Underground Technology Conference, Houston, Texas. Representatives of 15 utilities participated
in the review of the project goals, definitions, and the draft utility assessment protocol. Those
participating represent approximately 6 percent of the PCCP installed (by linear feet).
ASSESSMENT (EVALUATION) MATRICES
The goal of developing the PCCP assessment matrix is to allow a utility, utilizing its
existing data, to define the initial priorities for allocation of scarce resources in further assessing the
risk associated with its pipelines. The assessment matrix will become an initial screening tool, one
that can be maintained for each PCCP pipeline or pipeline segment within the utilitys system.
The basis of the matrix initially proposed during the utility workshop was related to the
critical nature of five components of PCCP pipeline success and what is known about those risk
components now:
Design as thorough as possible within the context of the state-of-the-art at that time
and either engineered by the owner or well reviewed by the owner
Manufacturing by a manufacturer who is committed to making each length of pipe
to the highest quality level attainable, consistent with the design
Inspection by both manufacturer and owner, in the plant and during construction
Construction by a contractor motivated to comply with the requirements of the
construction documents
Maintenance and Operations that do not place the pipeline at additional risk
The draft PCCP assessment matrix, in MS Excel format, is subdivided into five
submatrices as described above. Included are some of the more important factors within the
categories. The utility workshop resulted in significant refinement of these, as well as refinement
of the weighting factors, as described below.
PCCP Data Weighting Factors
In order to keep weighting factors simple, the qualitative risk-ranking matrix proposed
included only three rankings. They were intended to reflect the relative confidence in the data,
without considering the severity of the consequence. In other words, the qualities that are being
compared are compared on the equitable basis of knowledge. This allows the nuances of
particular characteristics of PCCP to not be overshadowed by the big picture. Only after
assembling the data and qualitatively assessing the sum of the confidence in the data can sound
risk decisions be made. To that end, the following three ranks were proposed to allow the
qualitative rating of any PCCP pipeline.
High Rank 1. The confidence in the data is high or the risk factor is not applicable
to the PCCP pipeline being assessed.
Medium Rank 2. The risk factor could be present or applicable, but based on the
data, it cannot be confirmed or denied.
Low Rank 3. The risk factor is confirmed or the utility has experience with the risk
factor on the PCCP pipeline being assessed.
116
Please note that the intent was to score each item on the basis above, so that even a
nonapplicable item still would rank 1. The intent was to establish through truthing of the
matrix by the participating utilities, a range of total scores that could be condensed into three
ranges of low, medium, and high confidence in the PCCP pipeline being assessed.
In the workshop, the participating utilities were asked what in their opinions the most
important factors were related to PCCP failures experienced within their systems. Those were:
Owners risk factors (for material): wire class and size (No. 1), manufacturer (No. 2),
slurry under wire (No. 3). Also cited were coating cast vs. sprayed, cylinder
thickness, age or date of manufacture (birth), presence of electrical bonding, quality
of external coating, plant vs. site manufactured pipe, and core quality.
Inspection and installation factors: inspection and records (No. 1), contractor (No. 2),
pressure class, quality of backfill, handling/transportation, joint mortaring/diapering,
and electrical continuity. Also cited was the presence of physical preloading (or
tension in the wire) actually present in the wire.
Environmental/operations factors: surge (No. 1), soil (No. 2), cathodic
protection/interference, fluctuating groundwater, surcharging (right-of-way)
management, appurtenance maintenance, ground movement, internal corrosion of
core, and underground to aboveground transition (aerials.)
Maintenance factors: post-construction inspection and monitoring (No. 1), data
analysis/study, life extension (rehab/repair), and design re-analysis.
The workshop participants comments on the Initial Risk-Based Assessment Matrix can
be summarized as follows:
The assessment factors would be better understood written as questions.

Each question should be written such that positive = less risk.
Answers (and score) to be yes, no, dont know, such that lack of knowledge about a
factor does not increase the risk. (After all, the assessment does not affect the risk
either.)
Possible scoring of answers: yes = 1, no = -1, dont know = 0.
Length of pipeline needs to be factored in, assuming that as the size of the project
increases, there is less effect on the learning curve of the manufacturer.
The participants stressed that in many instances, although it was clear that many factors
influence the longevity of PCCP, there is limited information available for many pipelines. Poor
record keeping is the norm, and reconstructing the information even when available is timeconsuming. The participants requested that a two-tier approach be developed: one quickly
determined assessment and another more extensive initial assessment based upon acquisition and
review of available data. The intent is to be able to determine what if anything needs to be
evaluated for any particular pipeline constructed of PCCP. The methodology is equally
applicable to LC-PCCP and EC-PCCP.
117
Short-Form Assessment Matrix

The short-form assessment matrix addresses the top five factors that are both likely to be
known about a PCCP pipeline and which have a significant effect on the pipeline. Scoring of
answers should be: yes = 1, no = -1, dont know = 0. Thus, if a pipeline scores a negative total it
may be prudent to evaluate it further.
Question 1.
Question 2.
Question 3.
Question 4.
Question 5.
Was the pipeline constructed prior to 1964 or after 1992? This question
relates to the higher statistical probability of PCCP constructed prior to
1964 (and the Principal Investigators have considered limiting this range
to prior to 1955) and the failure of only one PCCP pipeline constructed
since 1992.
Is the pipe manufacturer still in business on the site at which the pipe was
made? The intent of this pipeline may infer that pipe made by a certain
company, assailed by the surviving manufacturers as the source of the
troubles, may be condemned as a class. That is not so. Rather, the
implication is that a manufacturer still in business on-site may have
pursued a more conservative approach to manufacture and may have
access to construction records necessary for any further assessment of the
pipeline.
Is the pipe constructed with 6-gauge (0.192-inch diameter) or larger wire?
The highly stressed 8-gauge wire, although not solely the source of all
wire breaks, but it is vulnerable. So score this -1 if the pipe was
constructed with 8-gauge (0.162-inch diameter) wire.
Was inspection done at the manufacturing site and at the time of
installation by the owner or the owner's consulting engineer? There is no
substitute for inspection. If there is no inspection, the likelihood of
required tests being conducted or bedding placed as specified or the joints
being properly mortared is speculative at best. Further, inspection records
can be a great source of comfort should they be extant and confirm
manufacturing and installation in conformance with the intent of the
design.
Is surge unlikely on the pipeline? Score -1 if pumped flow or other
condition where surge is likely. The longevity of PCCP is significantly
lessened when subject to variations in application of load. And surge
pressures inadvertently applied in excess of the design curve tend to crack
the coating, exposing the wires to moisture and subsequent corrosion.
118
Long-Form Assessment Matrix

The longer form assessment matrix addresses many of the factors that are both likely to
be known (or can be discovered with examination of records) about a PCCP pipeline and which
have a significant effect on the pipeline. Scoring of answers should be: yes = 1, no = -1, dont
know = 0. The zero or dont know factor is intended to avoid skewing an assessment
without justification. Thus, if a pipeline scores a negative total it may be prudent to evaluate it
further. Factors that may indicate an appropriate design basis:
Question 1.
Question 2.
Question 3.
Did the project specification include requirements in addition to AWWA

Standard C301 or the USBR standard specification? Many designers
specified factors in excess of those minimums stated in either the
consensus standards or the manufacturers suggested specifications. The
assumption here is that an additional specification may have resulted in a
slightly more conservative pipe design or manufacture. It should also be
noted that the citation of the standards, particularly in more recent
editions, requires the specification of multiple choices for items that can
potentially affect the longevity of the product. A pipe order that was only
AWWA C301 was likely to be limited in those items to whatever the
manufacturer chose to include.
Did the project contract documents include minimum design requirements
(wire size and spacing, etc.)? Score this -1 if the pipe manufacturer
submitted the design. The assumption here is that a design by the
manufacturer included no excess of conservatism.
Was the design bedding angle less than or equal to 90 degrees? This is
recognizing that the design practice using either Paris or Olanders
coefficients to assume a 90-degree bedding angle in design, but that if
excess design moments and thrusts could be reduced by assuming 120
degrees then it was done. In actual practice, the shaping of bedding to
achieve 90-degree bedding requires extra effort and is a requirement
oftentimes ignored. A design based upon 120-degree bedding is at greater
risk. Score this -1 if 120-degree bedding angle or greater was assumed.
Factors that may indicate an appropriate manufacturing quality:

Question 4.
Question 5.
Is the pipe manufacturer still in business? The intent of this pipeline may
infer that pipe made by a certain company, assailed by the surviving
manufacturers as the source of the troubles, may be condemned as a class.
That is not so. Rather, the implication is that a manufacturer still in
business on-site may have pursued a more conservative approach to
manufacture and may have access to construction records necessary for
any further assessment of the pipeline.
Was the pipeline constructed prior to 1964 or after 1992? This question
relates to the higher statistical probability of PCCP constructed prior to
1964 (and the Principal Investigators have considered limiting this range
to prior to 1955) and the failure of only one PCCP pipeline constructed
since 1992.
119
Question 6.
Question 7.
Question 8.
Question 9.
Question 10.
Question 11.
Question 12.
Question 13.
Was the pipe manufactured in a fixed plant? Score this -1 if the pipe was
manufactured on-site. Pipe manufactured in a fixed plant is less subject to
the effect of weather or a transient workforce on the quality of the product.
Is the prestressing wire Class I or Class II? Score this -1 if Class III or
Class IV. Class II and the so-called Class IV wire were stressed higher
than the Class I and II wires by design.
Is all the reinforcing wire greater than No. 6? Score this -1 if No. 8 wire
was used. The highly stressed 8-gauge wire, although not solely the source
of all wire breaks, is vulnerable. So score this -1 if the pipe was
constructed with 8-gauge (0.162-inch diameter) wire.
Is the steel cylinder 16 gauge or thicker? Score this -1 if 18-gauge
cylinders are used. The thin cylinders were welded together with lap
joints to avoid burning through by butt-welding. Those weldments create
bending moments in the cylinder when the pipe is under pressure. The thin
cylinders were also more difficult to weld on the spigot and bell rings
without burn-through.
Is the design wire stress ratio less than 70 percent? Score this -1 if 75
percent of ultimate strength. Wire stressed at lower ratio of the ultimate
strength has more reserve capacity for vagaries in manufacturing and
operational events (such as sudden valve closure surges.)
Is the coating, cast or mortar, at least 7/8-inch thick? Thin coatings
provide less resistance to chloride penetration than do thicker, denser
coatings.
Was mortar slurry placed under the prestressing wires? This was required
after 1984, so if the date of manufacture is post 1984 and the pipe was
specified to the 1984 or later standard, the score should be +1.
Were bonding straps placed under wires to allow corrosion monitoring?
Score this +1 if shorting cables were installed between pipe segments.
Shorting straps allow the pipeline to be monitored for corrosion. It can be
reasonably assumed that a pipeline owner that paid for shorting straps is
motivated to monitor the pipeline.
Factors which may indicate appropriate inspection was done:

Question 14. Were pipe design submittals provided for review by owner? It can be
reasonably assumed that a pipeline owner that did a careful review of the
shop submittals got a better product manufactured.
Question 15. Was documentation of inspection and testing by manufacturer submitted
to the owner for review? It can also be reasonably assumed that a pipeline
owner that actually checked that required tests were done got a better
product manufactured.
120
Question 16. Was any manufacturing inspection done by the owner or owner's
engineer? There is no substitute for inspection. If there is no inspection,
the likelihood of required shop tests being conducted is speculative at best.
Further, inspection records can be a great source of comfort should they be
extant and confirm manufacturing and installation in conformance with
the intent of the design.
Question 17. Did the owner or owners engineer do any on-site construction inspection?
Again (and it needs to be repeatedly stressed), there is no substitute for
inspection. If there is no inspection, the likelihood of required compaction
tests being conducted or bedding placed as specified or the joints being
properly mortared is speculative at best. Further, inspection records can be
a great source of comfort should they be extant and confirm
manufacturing and installation in conformance with the intent of the
design.
Question 18. Were wire tests done in accordance with the then-current standards (i.e.,
mill certificates) and submitted to the owner for review? This also
presumes that pipe was not supplied from manufacturers stock.
Question 19. Did the pipe manufacturer provide on-site assistance to the installation
contractor? Projects benefit from hands-on assistance from the
manufacturers instructions on-site to the installation contractors field
superintendent. So it can be assumed that projects that proceeded with
installation without that assistance may have experienced difficulties
resulting in damage during that installation.
Factors which may indicate that appropriate care was taken during construction:
Question 20. Was the pipe placed on shaped bedding? This is recognizing that the
design practice using either Paris or Olanders coefficients to assume a
90-degree bedding angle in design, but that if excess design moments and
thrusts could be reduced by assuming 120 degrees then it was done. In
actual practice, the shaping of bedding to achieve 90-degree bedding
requires extra effort and is a requirement oftentimes ignored. A design
based upon 120-degree bedding is at greater risk. Score this -1 if 120degree bedding angle or greater was assumed.
Question 21. Was imported material utilized for bedding? Score this -1 if native soils
were used for bedding. If native soils are clean sand then this can be
scored +1.
Question 22. Was the hydrostatic test passed on the first attempt? A failure during
hydrostatic testing may otherwise be indicative of rolled gaskets. Other
failures during hydrostatic testing have resulted due to misplaced pipe.
That may indicate that other pipe were misplaced and are operating at
higher stress levels than intended.
121
Question 23. Was the hydrostatic test pressure no more than 20 percent over the design
operating pressure? AWWA Manual M-9 indicates a maximum 120
percent increase over the design working pressure for the field hydrotest.
In no case should the pipe at any point be hydrotested in excess of the
upper bound of the combined stress design curve of Appendix A or B in
AWWA C301. Unfortunately, the standard of practice in the waterworks
industry is to test distribution pipelines at 150 percent of the working
stress. That practice applied to PCCP pipelines results in permanently
cracked coating.
Question 24. Were joint bonds installed and tested? Shorting straps installed in the pipe
(see Question 12) are useless if the bond straps between pipes were not
installed and tested.
Question 25. (If yes to the above question, otherwise score this 0) Is the pipeline
electrically isolated from other pipelines? This is to check if inadvertent
cathodic protection (stray current) is being applied to the pipe. If so, it
should be scored 1 because of the potentially detrimental effect on the
prestressing wires. Further investigation in the field may be warranted if
the score is 1.
Question 26. Were joints welded for thrust restraint? Score this -1 if bolted joint
restraints, which allow slight movement, were used. That movement
would tend to open up circumferential cracks in the mortar coating and
lining at the joints, leading to exposed steel to water.
Question 27. Were joints mortared inside and out? Inspection records may be necessary
to make this determination, on a pipe-by-pipe basis.
Factors that may indicate appropriate operation and maintenance:
Question 28. Is surge unlikely on the pipeline? Score this -1 if pumped flow or other
condition where surge is likely or control valves exist on the pipeline
which could be accidentally closed, recognizing that the longevity of
PCCP is significantly lessened when subject to variations in application of
load. And surge pressures inadvertently applied in excess of the design
curve tend to crack the coating, exposing the wires to moisture and
subsequent corrosion.
Question 29. Is the earth fill over the pipeline substantially unchanged from when it was
constructed? Additional fill placed over a PCCP pipeline when not
addressed in design (such as a highway constructed on top of the
alignment) may result in detrimental cracking of the mortar coating or
concrete core. An alignment evaluation may be necessary to accurately
score this item.
Question 30. Is the pipeline subject to regular corrosion monitoring? It is assumed that a
pipeline that is monitored is likely to be in better condition than one that is
not.
122
Question 31. If cathodic protection is applied, is the voltage applied < 850 mv? Score
this -1 if > 1,000 mv and score it 0 if between these values or no cathodic
protection has been applied to the pipeline. Tests done in a laboratory
setting have demonstrated that cathodic protection applied below these
limits if not exceeded are unlikely to cause hydrogen embrittlement of the
prestressing wire.
Other Factors which may indicate condition:
Question 32. The pipeline has experienced neither a leak nor a break. Score this -1 if it
has leaked, etc., thus assessing the known failures appropriately as
failures.
Question 33. If an internal inspection has been done (of any type) there is little or no
indication of damage/cracks/spalls/broken wires. Score this -1 if
damage/cracks/spalls/broken wires are indicated. No attempt here is made
to assess the quality of the other assessment data or conclusions, merely
that if any indication of damage or defect are indicated. The evaluation of
those data are beyond the limits of this assessment.
Question 34. If soil corrosivity tests were performed, were all laboratory saturated soil
resistivities greater than 1,000 ohm-cm or field resistivities greater than
1,500 ohm-cm found at pipe depth? If no soil corrosivity tests were
performed, score this zero. It is generally understood that PCCP in soil
environments less resistive than these limits are at greater risk of
corrosion.
Question 35. If soil chemistry tests were performed, were all chloride levels less than
350 ppm (mg/kg)? If no soil chemistry tests were performed, score this
zero. It is generally understood that PCCP in soil environments more
salty than this limit are at greater risk of corrosion.
Question 36. If soil chemistry tests were performed, were pH values greater than 5.5?
Again, if no soil chemistry tests were performed, score this zero. Acidic
soils with pH less than 5.5 will attack the mortar or concrete coating used
to protect the prestressing wire.
Question 37. If close interval pipe-to-soil potentials have been measured, are any areas
greater than 50 feet in length more negative than -300 millivolts to coppercopper sulfate? If no soil corrosivity tests were performed, score this zero.
This test can discern limited lengths of pipe at risk to corrosion due to
changes in the electropotential differences in the soils environment.
Question 38. Groundwater levels are stable. Score this -1 if any portion of the pipeline
is in an area of fluctuating groundwater levels. The fluctuating
groundwater leads to wetting-drying cycles that can lead to corrosion of
the wires.
Repeating the instructions to the initial assessment process, scoring of answers should be:
yes = 1(positive one), no = -1 (negative one), dont know = 0 (zero, or no score). Thus, if a
pipeline scores a negative total, it may be prudent to evaluate it further.
123
VALIDATION OF ASSESSMENT MATRICES

The final assessment forms were utilized by participating utilities to evaluate 22
pipelines. Of those 22, nine short forms were also completed, allowing comparison. Short forms
total scores ranged from -5 to 0 for pipelines that had experienced a failure. Long forms total
scores ranged from -9 to +24 for pipelines that had experienced a failure. It is clear then that for
use of the short form, any net score less than +1 should be a candidate for further evaluation.
Examining the scores of the long-form assessments it became clear that those that scored
below zero had multiple problems. What was most curious about those that had high scores and
still experienced failures was that the common thread was excessive surge pressures.
Examination of the contract documents of those also revealed little evidence of design
consideration for the surge or the hydrostatic test as design conditions. Those two items are red
flags waving. Clearly, a PCCP pipeline can have one or two negatives against it and fail, for the
weight of the positives may not override the weaknesses in the system.
The checklists are thus not intended to be a definitive failure predictor, rather part of a
multiphased approach. The intent is to provide a tool to the PCCP-owning utility to do a
preliminary self-assessment of its pipelines (using the short form) followed, if indicated, by a
more refined self-assessment. It was not uncommon for utilities to include with their assessments
comments similar to One thing this showed us is that we don't have nearly as much info as we
should. Carefully completed, it is intended that the assessment checklists, both long and short
form, will be useful as pipe inspection and maintenance prioritization tools and possible input to
risk assessment of PCCP pipelines.
124
CHAPTER 5
PREDICTING SERVICE LIFE
OGIVES AND FAILURE FORECASTING

The previous histograms and simple statistical analyses examined PCCP failures that had
already occurred. The next progression is to utilize these failure data in order to forecast likely
future occurrences.
One way to do this is by cumulative frequency distributions, usually expressed
graphically in a line graph called an ogive (pronounced 'jv'). Histograms visualize the existing
data. Ogive uses historical data to make predictions based only on what failed (no credit given
for the population that is still in service). The ogive plots cumulative frequency versus the upper
boundary of a class. The main use of an ogive is to calculate percentiles, or estimates, of the
proportion of data that falls above and below a certain value. Here, the ogive is used to plot the
cumulative frequency of PCCP failures as a function of age. This allows the determination of the
25th, 50th, and 75th percentiles (also called the 1st, 2nd, and 3rd quartiles or even the lower quartile,
median, and upper quartile). These values represent the number of years it took for 25, 50, and
75 percent of the pipe failures to occur. The value is obtained by linear interpolation between
upper class boundaries.
On the first ogive (Figure 5.1), the left y-axis, or ordinate, is the cumulative frequency of
all Category 1 failures and is represented by blue lines with blue diamonds. The right ordinate is
the cumulative frequency of all Categories 2 and 3 failures, illustrated by pink lines and boxes.
Age is incremented in five-year intervals along the x-axis, or abscissa. The distribution of both
datasets is S-shaped, indicative of normally distributed data. It can be seen that the slope of both
curves drastically decreases after 35 years. This equates with a reduction of failure frequency
after this time period. Now turning attention to forecasting, after 6.7 years of pipe age, 25
percent of the Category 1 failures had occurred. After 11.5 years, half of the Category 1 failures
had occurred. After 18.8 years, 75 percent of the Category 1 failures had occurred. Categories 2
and 3 failures, with a 10 to 15-year delayed response from the Category 1 failures, took 22.1
years, 26.3 years, and 29.4 years for 25 percent, 50 percent, and 75 percent of the Categories 2
and 3 failures to occur, respectively. Red leader lines depict where each of these values lie in
relation to the cumulative frequency for their respective dataset.
The next ogive, Figure 5.2, illustrates the cumulative frequency of Category 1 and
Categories 2 and 3 failures occurring in pipes installed before 1955. The left ordinate is the
cumulative frequency of Category 1 failures and is represented by blue lines with blue diamonds.
The right ordinate is the cumulative frequency of Categories 2 and 3 failures, illustrated by pink
lines and boxes. Age is again incremented in five-year intervals along the abscissa. The
distribution of both datasets is S-shaped, indicative of normally distributed data. It can again be
seen that the slope of both curves drastically decreases after 35 years, or approximately the year
1983. This equates with a reduction of failure frequency after this time period. Looking at
forecasting, it was found that it took 19.0 years, 23.5 years, and 28.6 years for 25 percent, 50
percent, and 75 percent of the Category 1 failures to occur, respectively. As for Categories 2 and
3 failures, it took 21.3 years, 23.3 years, and 26.3 years for 25 percent, 50 percent, and
125
75 percent of the Categories 2 and 3 failures to occur, respectively. This represents less than a
two-year gap between Category 1 and Categories 2 and 3 failures. At the 25-year mark,
Categories 2 and 3 failures even lead the Category 1 failures. This is indicative that the
Categories 2 and 3 failures in this timeframe were more independent failures then reactions to
Category 1 failures.
The next ogive, as found in Figure 5.3, depicts the cumulative frequency of Category 1
and Categories 2 and 3 failures for pipelines installed between 1955 and 1963. The axes are as
described above. Both curves have changes in slope occurring at 40 years, indicative of
deviations from normal distribution. The constant zero-value of Categories 2 and 3 failures for
the first 20 years, followed by a large change in slope further indicates a non-normal distribution.
It was found that it took 18.0 years, 23.6 years, and 40.6 years for 25 percent, 50 percent, and 75
percent of the Category 1 failures to occur, respectively. The 25th and 50th percentiles for this era
were nearly identical that of previous timeframe, however, the 75th percentile for this era
occurred 12 years later than the last eras. This indicates that failures were occurring over a
longer stretch of time in this timeframe. As for Categories 2 and 3 failures, it took 22.2 years,
24.5 years, and 41.5 years for 25 percent, 50 percent, and 75 percent of the Categories 2 and 3
failures to occur, respectively. Again for the Categories 2 and 3 failures, the 25th and 50th
percentiles for this era were similar to the last eras. Here again also, the 75th percentile occurs
after a longer time lapse, in this case over 15 years than the previous timeframe.
Figure 5.4, the next ogive, illustrates the cumulative frequency of failures occurring in
pipe installed in 196467. Present time is being approached in this graph; the grayed out area
between 45 and 50 years represents future time. A change in slope in the Category 1 failure
curve at 10 years and in the Categories 2 and 3 failure curve at 30 years are indicative of
deviation from normal distribution. It was found that it took 10.9 years, 12.9 years, and 14.8
years for 25 percent, 50 percent, and 75 percent of the Category 1 failures to occur,
respectively. This represents a decrease of 7.1 years, 10.7 years, and 25.8 years for the
corresponding percentiles of the last timeframe. For Categories 2 and 3 failures, it was found
that it took 11.6 years, 17.7 years, and 30.1 years for 25 percent, 50 percent, and 75 percent of
the failures to occur, respectively. These correspond with a decrease of 10.6 years and 6.8
years for the 25th and 50th percentile of the last eras, and an increase of 15.2 years for the 75th
percentile. This indicates that more failures occurred in a shorter amount of time in the
beginning, but later failures occurred over a longer stretch of time when compared to those
pipes installed in the previous era.
Figure 5.5, the lifespan ogive for pipe installed between 1968 and 1971, represents the
cumulative frequency of failures for this era. Present time is again being approached in this
graph; the grayed out area between 40 and 50 years represents future time. The Category 1
failure curve is relatively S-shaped, indicative of normally distributed data. The Categories 2 and
3 failure curve has a change in slope at 25 years and a constant slope between 5 and 15 years,
both indicators of a non-normal distribution. Turning to the forecast, it took 6.2 years, 8.4 years,
and 11.5 years for 25 percent, 50 percent, and 75 percent of the Category 1 failures to occur,
respectively. This represents a decrease of 35 years for the corresponding percentiles of the last
timeframe. For Categories 2 and 3 failures, it was found that it took 8.4 years, 12.5 years, and
26.3 years for 25 percent, 50 percent, and 75 percent of the failures to occur, respectively. These
also correspond with a decrease of 35 years for the 25th, 50th, and 75th percentiles of the last era.
126
The next ogive, depicted as Figure 5.6, show the cumulative frequency of all failures for
pipelines installed in 197278. Both the Category 1 and Categories 2 and 3 failure curves are
relatively S-shaped, indicative of normally distributed data. The Categories 2 and 3 failure curve
is more elongated in the 2535 timeframe. Turning to the forecast, it took 6.1 years, 9.9 years,
and 16.9 years for 25 percent, 50 percent, and 75 percent of the Category 1 failures to occur,
respectively. This represents an increase of 05 years for the corresponding percentiles of the
last timeframe, reversing the trend of the last three eras. For Categories 2 and 3 failures, it was
found that 25.1 years, 27.6 years, and 30.2 years passed for 25 percent, 50 percent, and 75
percent of the failures to occur, respectively. These values correspond with an increase of 16.7
years, 15.1 years, and 3.9 years respectively for the 25th, 50th, and 75th percentile of the last era,
again a reversal of the trend observed over the last three eras.
The next two figures, Figure 5.7 and 5.8, illustrate the breakdown of cumulative failures
in the 197278 era between Interpace Corporation pipe, and unknown/non-Interpace Corporation
pipe. As far as sample populations, there was nearly a fourfold increase of Interpace-related
failures versus unknown/non-Interpace failures in this timeframe. Both the Interpace and
unknown/non-Interpace Category 1 failures followed an S-curve shape, indicating normally
distributed data.
The Categories 2 and 3 failures described in Figures 5.7 and 5.8 for both Interpace and
unknown/non-Interpace had changes in slope that point toward behavior of non-normally
distributed data. Examining the forecast, it was found that it took 5.5 years, 9.1 years, and 14.7
years for Interpace pipe and 9.3 years, 16.9 years, and 24.2 years for unknown/non-Interpace
pipe for 25 percent, 50 percent, and 75 percent of the Category 1 failures to occur, respectively.
This represents a 4- to 10-year difference between the Interpace and unknown/other
manufactured pipe. For Categories 2 and 3 failures, it was found that it took 21.4 years, 23.7
years, and 26.9 years for Interpace pipe and 26.4 years, 28.6 years, and 31.2 years for
unknown/non-Interpace pipe for 25 percent, 50 percent, and 75 percent of the failures to occur,
respectively. These represent a five-year difference between the Interpace and other
manufactured pipe. In every percentile, Interpace pipe failed sooner than its unknown/nonInterpace counterpart. A distinct increase in both Interpace and unknown/non-Interpace
Categories 2 and 3 pipe failures occurred in the 20- to 25-year range.
The following ogive, Figure 5.9, represents the cumulative Category 1 and Categories 2
and 3 failures for the 197991 data partition. The distributions of both datasets are irregular,
indicative of non-normally distributed data. For Category 1 failures, the large constant curve
indicates a J-shaped distribution. For the Categories 2 and 3 failure curve, the change in slope at
20 years followed by large constant line indicates a skewed J-shaped curve. As far as
forecasting, it took 2.9 years, 6.1 years, and 10.4 years for 25 percent, 50 percent, and 75 percent
of the Category 1 failures to occur, respectively. This is decrease of 36 years from the allsamples population of the last era. For Categories 2 and 3 failures, it took 21.3 years, 24.0 years,
and 26.9 years for 25 percent, 50 percent, and 75 percent of the Categories 2 and 3 failures to
occur, respectively. This represents a 34 year decrease from the corresponding percentiles from
the last era and a 1618 year difference between Category 1 and Categories 2 and 3 failures
during this era.
Figure 5.10 illustrates the cumulative frequency of failures for pipe installed between
1992 and 2007. Similar to the lifespan histogram for this era, this ogive is provided for
completenessstatistically, not enough failures have occurred to accurately forecast failures.
127
Both Category 1 and Categories 2 and 3 failure curves are irregularly shaped. As far as the
forecast, it took 11.3 years, 12.5 years, and 13.8 years for 25 percent, 50 percent, and 75 percent
for both the Category 1 and Categories 2 and 3 failures to occur. It should be stressed that these
values were obtained by interpolating over a single failure entry a pipeline that experienced a
Category 1 failure and was subsequently relined, and any inference of the results may be quite
erroneous.
Table 5.1 summarizes in tabular form all the data presented here.
WEIBULL ANALYSIS AND FAILURE PREDICTION
Initially, histograms and simple statistical analyses were used to examine failures that
had already occurred. Next, ogives were prepared that utilized these statistical data in order to
forecast likely future occurrences of failures based on trends and percentiles. The last step is to
incorporate unfailed pipe in order to predict future failure rates.
One way to do this is by Weibull (commonly pronounced 'wI'bul) probability
distribution, usually expressed graphically as unreliability, or probability of failure, versus time
on a log-log grapha graph with logarithmic scales on both axes. Weibull gives credit for
population still in service and allows prediction of failure probability as a function of age. In
order to adequately model the failure data, the three-parameter Weibull probability density
function was employed:
(T) =
where:
(T) = Weibull probability density function at time T
(Beta) = shape parameter or slope (dimensionless)
(Eta) = scale parameter (units of T)
(Gamma) = location parameter (units of T)
In order to determine how well each function fit the data, a correlation coefficient (, rho)
was used. For the three-parameter Weibull density function, a nonlinear regression analysis was
employed to fit the data to a curve rather than a line. The nonlinear model was then
approximated by linear terms and standard regression (least-squares method) used in order to
obtain the correlation coefficient. The closer the coefficient is to 1, the better the function fits
the data. A coefficient of zero indicates random data with no correlation to the function.
128
Table 5.1
Summary statistics
By using the Weibull distribution with suspensions (right censored data, or unfailed
samples), a more accurate model of life data analysis for PCCP was achieved. These unfailed
pipe were calculated by subtracting the known failures from the total production of pipe on a
year-by-year basis.
Weibull analysis was performed on each of the data partitions to obtain failure rates for
every era of PCCP pipe manufacturing. In this study, ReliaSofts Weibull++ 7 software was
used to compile the forthcoming charts. Other commercial software packages are available.
The first Weibull plot, Figure 5.11, utilizes all the available failure sample population and
suspensions (unfailed individual sticks of pipe). The y-axis, logarithmically scaled, is the
probability of Category 1 failures in percent failed per pipe produced. The x-axis, also
logarithmically scaled, represents the age of the pipeline in years. Using the Weibull parameters
and , the probability function was plotted in black diamond unadjusted points and black
unadjusted line.
An arbitrary location factor, , was selected and subsequently iterated to best fit the data
as a straight line. This can be seen on the plot as the blue round adjusted points and blue adjusted
line. A value of 0.9801 demonstrates a relatively good fit of the curve to the data points. It
should be noted that the last failures occur in the 45-year age range, points along the line to the
right of these failures represent the prediction of failures at a future time. Because this plot is
129
representative of the failure probability of all pipe manufactured between 1942 and 2006, points
before 45 years may also be used predicatively. The failure rate probability at 10 years and 100
years was 2.82 x 10-03 and 0.03 percent, respectively.
The next Weibull plot, Figure 5.12, examines the failure rate prediction for Categories 1
and 2 failures for all samples. Adding the Category 2 failures effectively doubled the failure
sample population, to a total of 758 failures. The number of suspensions, or unfailed pipe,
remained the same at 4,951,634 pipes. The logarithmically scaled y-axis is the probability of
Categories 1 and 2 failures in percent failed per pipe produced. The x-axis represents the age of
the pipeline in years and is also logarithmically scaled. The unadjusted probability function was
plotted in black diamond points and the black unadjusted line. Using the location factor, the
function was adjusted on the plot as illustrated by the blue round points and the blue line. A of
0.9832 demonstrates relatively good fit of the curve to the data points and is also a slightly better
fit than the last Weibull plot of just the Category 1 failures. Again, because this sample
population represents all pipe manufactured between 1942 and 2006, any point on the function
may be used predicatively. The Categories 1 and 2 probability of failure at 10 and 100 years was
4.62 x 10-03 and 0.07 percent, respectively.
Next, Figure 5.13 looks at the failure rate for Categories 1, 2, and 3 failures for all
samples. Adding Category 3 failures increased the failure sample population 33-fold, for a total
of 25,217 failures. The number of suspensions, or unfailed pipe, still remained the same at
4,951,634 pipes. The logarithmically scaled y-axis is the probability of Category 1 and
0.9535 demonstrates a decent fit of the curve to the data points but is lower than those obtained
for the Category 1 and Categories 1 and 2 plots. Graphically, this can be seen as more data points
deviating from the path of the curve. Once again, because this sample population represents all
pipe manufactured between 1942 and 2006, any point on the function may be used predicatively.
The Category 1 and Categories 2 and 3 probability of failure at 10 years and 100 years was 0.03
and 3.38 percent, respectively.
Figure 5.14, the next Weibull plot, looks at just the Category 3 failures from 1942 to
2006. The failure sample population was 24,459. The number of suspensions, or unfailed pipe,
still remained the same at 4,951,634 pipes. The logarithmically scaled y-axis is the probability of
Category 3 failures in percent failed per pipe produced. The x-axis represents the age of the
pipeline in years and is also logarithmically scaled. The unadjusted probability function was
0.9659 demonstrates a decent fit of the curve to the data points but is lower than those obtained
for the Category 1 and Categories 1 and 2 plots. The Category 3 probability of failure at 10 and
100 years was 0.03 and 23.75 percent, respectively.
Simply put, given all PCCP already produced and those already failed, it is predicted that
any particular remaining stick of PCCP has a 0.03 percent probability of rupturing within 100
years of its installation and a 3.38 percent probability of experiencing some other failure or
replacement within the same 100 years. That said, PCCP pipe produced in some eras is more or
130
less susceptible to failure than others, as previously discussed in this report. More accurate
predictions can be made by narrowing down the timeframe that the PCCP was installed. These
more refined predictions follow.
Figure 5.15, the next Weibull plot, looks at Category 1 failures from PCCP installed prior
to 1955. The failure sample population was 32. The number of suspensions, or unfailed pipe, is
476,376 pipes. The logarithmically scaled y-axis is the probability of Category 1 failures in
percent failed per pipe produced. The x-axis represents the age of the pipeline in years and is
also logarithmically scaled. The unadjusted probability function was plotted in black diamond
points and the black unadjusted line. Using the location factor, the function was adjusted on the
plot as illustrated by the blue round points and the blue line. A of 0.9931 demonstrates an
excellent fit of the curve to the data points, as can be seen graphically by how the data points are
grouped closely around the curve. The Category 1 probability of failure at 10 and 100 years was
Next, Figure 5.16 looks at the failure rate for Categories 1 and 2 failures from PCCP
installed prior to 1955. The number of suspensions, or unfailed pipe, still remained the same at
476,376 pipes. The logarithmically scaled y-axis is the probability of Categories 1 and 2 failures
in percent failed per pipe produced. The x-axis represents the age of the pipeline in years and is
plot as illustrated by the blue round points and the blue line. A of 0.9865 demonstrates a good
fit of the curve to the data points. The Category 1 and Categories 2 and 3 probability of failure at
10 years and 100 years was 2.00 x 10-02 and 0.37 percent, respectively.
The next Weibull plot, Figure 5.17, examines the failure rate prediction for Category 1
failures for PCCP installed in 195563. The sample population was 40 failures. The number of
suspensions, or unfailed pipe, was 1,049,061 pipes. The logarithmically scaled y-axis is the
probability of Category 1 failures in percent failed per pipe produced. The x-axis represents the
age of the pipeline in years and is also logarithmically scaled. The unadjusted probability
function was plotted in black diamond points and the black unadjusted line. Using the location
factor, the function was adjusted on the plot as illustrated by the blue round points and the blue
line. A of 0.9829 demonstrates relatively good fit of the curve to the data points and is also a
slightly better fit than the Weibull plot of all Category 1 failures. The Category 1 probability of
failure at 10 and 100 years was 2.32 x 10-04 and 0.02 percent, respectively.
Figure 5.18, the next Weibull plot, looks at the Categories 1 and 2 failures from 1955 to
1963. The failure sample population was 73. The number of suspensions, or unfailed pipe,
remained the same at 1,049,061 pipes. The logarithmically scaled y-axis is the probability of
0.9896 demonstrates a good fit of the curve to the data points and is higher than that obtained for
the Category 1 failures for the same era. The Categories 1 and 2 probability of failure at 10 and
100 years was 2.83 x 10-04 and 0.03 percent, respectively.
Next, Figure 5.19 looks at the failure rate for Categories 1, 2, and 3 failures from PCCP
installed in 1955 to 1963. The number of failed pipe and suspensions was 2,422 failures and
1,049,061 pipes, respectively. The logarithmically scaled y-axis is the probability of Category 1
131
and Categories 2 and 3 failures in percent failed per pipe produced. The x-axis represents the age
of the pipeline in years and is also logarithmically scaled. The unadjusted probability function
was plotted in black diamond points and the black unadjusted line. Using the location factor,
the function was adjusted on the plot as illustrated by the blue round points and the blue line. A
of 0.8605 demonstrates a much worse fit of the curve to the data points as compared to the
Category 1 or Categories 1 and 2 plots in the same era. The Category 1 and Categories 2 and 3
probability of failure at 10 and 100 years was 2.07 x 10-04 and 12.21 percent, respectively.
To verify that by including Category 3 failures in the previous plot worsened the
prediction, Category 3 failures for 195563 were plotted alone in the next Weibull plot, Figure
5.20. The failure sample population was 2,349, the bulk of the failures presented in the previous
plot. The number of suspensions remained the same at 1,049,061 pipes. The logarithmically
scaled y-axis is the probability of Category 3 failures in percent failed per pipe produced. The xaxis represents the age of the pipeline in years and is also logarithmically scaled. The unadjusted
probability function was plotted in black diamond points and the black unadjusted line. Using
the location factor, the function was adjusted on the plot as illustrated by the blue round points
and the blue line. A of 0.8272 verified the poor fit of the curve to the data points. The Category
3 probability of failure at 10 and 100 years was 0.09 and 0.72 percent, respectively.
Figure 5.21, the next Weibull plot, looks at Category 1 failures from PCCP installed in
was 594,178 pipes. The logarithmically scaled y-axis is the probability of Category 1 failures in
plot as illustrated by the blue round points and the blue line. A of 0.9707 demonstrates a good
fit of the curve to the data points. The Category 1 probability of failure at 10 and 100 years was
The next Weibull plot, Figure 5.22, examines the failure rate prediction for Categories 1
and 2 failures for PCCP installed between 1964 and 1967. Adding the Category 2 failures more
than doubled the failure sample population, to a total of 81 failures. The number of suspensions
remained at 594,178 pipes. The logarithmically scaled y-axis is the probability of Categories 1
and 2 failures in percent failed per pipe produced. The x-axis represents the age of the pipeline in
years and is also logarithmically scaled. The unadjusted probability function was plotted in black
diamond points and the black unadjusted line. Using the location factor, the function was
adjusted on the plot as illustrated by the blue round points and the blue line. A of 0.9643
demonstrates a decent fit of the curve to the data points but worse than that obtained for
Category 1 failures alone for the same era. The Categories 1 and 2 probability of failure at 10
and 100 years was 3.57 x 10-03 and 0.04 percent, respectively.
Next, Figure 5.23 looks at the failure rate for Category 1 failures from PCCP installed in
196871. The number of failures and suspensions was 60 failures and 551,218 pipes,
respectively. The logarithmically scaled y-axis is the probability of Category 1 failures in
points and the black unadjusted line. The function was adjusted on the plot as illustrated by the
blue round points and the blue line by using the location factor. A of 0.9429 demonstrates a
decent fit of the curve to the data points. The Category 1 probability of failure at 10 years and
132
remained at 551,218 pipes. The logarithmically scaled y-axis is the probability of Categories 1
demonstrates a decent fit of the curve to the data points; better than that obtained for the
Category 1 plot in the same era alone. The Categories 1 and 2 probability of failure at 10 and 100
suspensions was 838,154 pipes. The logarithmically scaled y-axis is the probability of Category
1 failures in percent failed per pipe produced. The x-axis represents the age of the pipeline in
demonstrates relatively good fit of the curve to the data points and is also a slightly better fit than
any of the Weibull plots of the last two eras. The Category 1 probability of failure at 10 and 100
Next, Figure 5.26 looks at the failure rate for Categories 1 and 2 failures from PCCP
installed in 197278. The number of failures and suspensions was 350 failures and 838,154
pipes, respectively. The logarithmically scaled y-axis is the probability of Categories 1 and 2
failures in percent failed per pipe produced. The x-axis represents the age of the pipeline in years
and is also logarithmically scaled. The unadjusted probability function was plotted in black
diamond points and the black unadjusted line. The function was adjusted on the plot as
illustrated by the blue round points and the blue line by using the location factor. A of 0.9790
demonstrates a good fit of the curve to the data points and slightly better than the Category 1
failures in this era alone. The Categories 1 and 2 probability of failure at 10 years and 100 years
was 0.01 and 0.21 percent, respectively.
Figure 5.27 looks at the failure rate for Categories 1, 2, and 3 failures for the 197278
era. Adding Category 3 failures increased the failure sample population 44-fold, for a total of
15,354 failures. The number of suspensions, or unfailed pipe, still remained the same at 838,154
pipes. The logarithmically scaled y-axis is the probability of Category 1 and Categories 2 and 3
failures in percent failed per pipe produced. The x-axis represents the age of the pipeline in years
and is also logarithmically scaled. The unadjusted probability function was plotted in black
demonstrates a decent fit of the curve to the data points but is the lowest of all the plots in this
era. Graphically, this can be seen as more data points deviating from the path of the curve. The
Category 1 and Categories 2 and 3 probability of failure at 10 years and 100 years was 0.08 and
4.01 percent, respectively.
133
Figure 5.28, the next Weibull plot, looks at just the Category 3 failures from 1972 to
1978. The failure sample population was 15,004. The number of suspensions, or unfailed pipe,
still remained the same at 838,154 pipes. The logarithmically scaled y-axis is the probability of
Category 3 failures in percent failed per pipe produced. The x-axis represents the age of the
pipeline in years and is also logarithmically scaled. The unadjusted probability function was
0.9661 demonstrates a decent fit of the curve to the data points. The Category 3 probability of
failure at 10 and 100 years was 0.05 and 100 percent, respectively. This alarming probability of
failure is a function of the nature of Category 3 failures. More of this is discussed in the results
section of this report.
suspensions, or unfailed pipe, was 1,061,650 pipes. The logarithmically scaled y-axis is the
probability of Category 1 failures in percent failed per pipe produced. The x-axis represents the
age of the pipeline in years, and is also logarithmically scaled. The unadjusted probability
function was plotted in black diamond points and the black unadjusted line. Using the location
factor, the function was adjusted on the plot as illustrated by the blue round points and the blue
line. A of 0.9398 demonstrates a decent fit of the curve to the data points. The Category 1
probability of failure at 10 and 100 years was 2.18 x 10-03 and 7.81 x 10-03 percent,
respectively.
1991. The failure sample population was 116. The number of suspensions remained at 1,061,650
pipes. The logarithmically scaled y-axis is the probability of Categories 1 and 2 failures in
plot as illustrated by the blue round points and the blue line. A of 0.9335 demonstrates a decent
fit of the curve to the data points. The Categories 1 and 2 probability of failure at 10 and 100
Next, Figure 5.31 looks at the failure rate for Categories 1, 2, and 3 failures from PCCP
installed in 1979 to 1991. The number of failed pipe and suspensions was 5,899 failures and
1,061,650 pipes, respectively. This represents a 50-fold increase of the failure sample
population. The logarithmically scaled y-axis is the probability of Category 1 and Categories 2
demonstrates a similar fit of the curve to the data points as compared to the Categories 1 and 2
plot for the same era. The Category 1 and Categories 2 and 3 probability of failure at 10 and 100
years was 0.05 and 1.36 percent, respectively.
Similar to the histogram and ogive for the 19922006 era, the Weibull plot for this time
period, Figure 5.32, is included only for completeness. A failure sample population of 1 is not
significant enough to make serious predictions. The number of suspensions for this era was
380,997 pipes. The logarithmically scaled y-axis is the probability of Categories 1 and 2 failures
in percent failed per pipe produced. The x-axis represents the age of the pipeline in years and is
134
also logarithmically scaled. A single sample population necessitated the use of a simpler twoparameter Weibull probability density function. The Category 1 probability of failure at 10 and
Likewise for Category 3 failures for the 19922006 era, Weibull plot Figure 5.33 is
included for entirety. A single failure event consisting of 1,299 pipes cannot accurately predict
PCCP failure rates for this era. The suspensions remained at 380,997 pipes. The single failure
event here also necessitated the use of a simpler two-parameter Weibull probability density
function. The Category 3 probability of failure at 10 and 100 years was 0.09 and 100 percent,
respectively. Although based on very limited statistics, this alarming probability of failure is
discussed in the results section of this report.
PROBABILITY OF PCCP FAILURE BY ERA
The next two graphs, Figures 5.34 and 5.35, summarize the predicted probabilities of
PCCP failures as a function of timeframe, or era of pipe manufacturing. The first bar graph
summarizes the probability of failure after 10 years; the second summarizes the probability of
failure after 100 years. The left-hand axis corresponds with the solid blue bars which represent
the probability of a Category 1 failure. The right-hand axis corresponds with the crossed
magenta bars which represent the probability of Category 1 or 2 failure. The pipe installation
timeframe increases along the bottom x-axis. Observing the probabilities trends in the first
graph, it can be seen that the largest Category 1 failure probability occurred in the 197278 era.
This seems to agree with the simple statistics derived earlier in this study. The highest
Categories 1 and 2 failure probability was bimodal, peaking both in the 197278 era and the pre1955 era.
On the 100-year probability chart, Figure 5.35, it can be seen that the Category 1 failure
probability trend was multimodal, with peaks in the 197278 era, pre-1955 era, and 19922006
era. If the 19922006 peak is ignored due to the aforementioned lack of sample size, the trend
becomes bimodal. The Categories 1 and 2 failure probability trend appeared bimodal around the
197278 era and the pre-1955 era, very similarly to the 10-year probability plot. Unlike the 10year probability plot and seemingly unintuitive when compared to the simple statistics derived
earlier in this study, the Category 1 failure probability does not peak in the 197278 era.
Table 5.2 summarized in tabular form all the Weibull data presented herein.
135
Table 5.2
Weibull summary statistics
136
Category 3 Failure Prediction Interference
It was noted that Category 3 failures (loss of service, i.e., full or partial replacement)
dramatically impacted the failure probabilities when included in Weibull prediction analysis.
Figure 5.18 vs. 5.19 is a keen example of the marked degradation of how the function fit the data
(i.e., curve fit) when adding Category 3 failures, as was Figure 5.26 vs. 5.27 and Figure 5.12 vs.
5.13.
This phenomenon is believed to be caused by the inherent severity of Category 3 failures.
When a PCCP pipeline alignment is partially or completely replaced, many pipes of as-yet
unfailed pipe are removed from service. This removal of many similarly manufactured pipes at
one particular time is not easily described by the Weibull probability function. An analogy
would be trying to calculate the failure probability of a product after a large nationwide recall
many unfailed products would unavoidably be counted as failed.
In an attempt to describe the rapid increase of failures, the Weibull function curve
increases dramatically. The result is large and unreliable predictions of failure probabilities.
Figures 5.14 and 5.19 saw double-digit probabilities of failure at the 100-year mark. Figures 5.27
and 5.33 observed alarming 100 percent failure probabilities at the 100-year mark.
To compensate for this occurrence, the Weibull analysis was performed for Categories 1
and 2 failures for all the manufacturing eras (except 1992-2006, which had no Category 2
failures). Category 2 failures, or failures by inspection, still accounted for nonruptured pipe
failures without counting the as-yet unfailed pipe. The results were a more plausible, credible
failure model.
137

Lifespan Ogive
All Samples
450
138
Cumulative frequency (no. of category 1 failures)
400
Category 1
th
25 percentile = 6.7 years
th
th
350
25,000
300
20,000
250
15,000
200
150
10,000
Category 2 & 3
th
th
th
100
50
5,000
0
0
10
15
20
25
Age (years)
Figure 5.1 Lifespan ogive all samples
30
35
40
45
50
Cumulative frequency (no. of category 2 & 3 failures)
30,000

Lifespan Ogive
Installed pre-1955
12
Category 1
th
th
th
139
30
25
10
8
20
6
15
4
Category 2 & 3
th
th
th
10
0
0
10
15
20
25
30
35
40
45
50
Age (years)
~1948
~1953
~1958
~1963
~1968
~1973
~1978
Figure 5.2 Lifespan ogive installed pre-1955
~1983
~1988
~1993
~1998
35

Lifespan Ogive
Installed 19551963
140
40
Category 1
th
th
th
35
2,500
2,000
30
25
1,500
20
1,000
15
Category 2 & 3
th
th
th
10
500
0
0
10
15
20
25
30
35
40
45
50
~1994
~1999
~2004
~2009
Age (years)
~1959
~1964
~1969
~1974
~1979
~1984
~1989
Figure 5.3 Lifespan ogive installed 1955-1963
3,000
45

Lifespan Ogive
Installed 19641967
70
Category 1
th
th
th
141
30
60
25
50
20
40
15
30
Category 2 & 3
th
th
th
10
20
10
0
0
10
15
20
25
30
35
40
~2001
~2006
45
50
Age (years)
~1966
~1971
~1976
~1981
~1986
~1991
~1996
~2011
~2016
35

Lifespan Ogive
Installed 19681971
50
70
45
40
50
35
Category 1
th
th
th
40
30
25
30
20
15
Category 2 & 3
th
th
th
20
10
10
5
0
0
10
15
20
25
30
35
40
45
~2005
~2010
~2015
50
Age (years)
~1970
~1975
~1980
~1985
~1990
~1995
~2000
~2020
142
60

Lifespan Ogive
Installed 19721978
16,000
Category 1
th
th
th
200
14,000
12,000
10,000
150
8,000
100
Category 2 & 3
th
th
th
50
6,000
4,000
2,000
0
0
10
15
20
25
30
35
40
~2010
~2015
45
50
Age (years)
~1975
~1980
~1985
~1990
~1995
~2000
~2005
~2020
~2025
143
250

Lifespan Ogive
Installed 19721978: Interpace
5,000
160
Category 1
th
th
th
120
4,500
4,000
3,500
100
3,000
2,500
80
2,000
Category 2 & 3
th
th
th
60
40
1,500
1,000
20
500
0
0
~1975
~1980
10
~1985
15
~1990
20
25
30
Age (years)
~1995
~2000
~2005
Figure 5.7 Lifespan ogive installed 1972-1978: Interpace
35
~2010
40
~2015
45
~2020
50
~2025
144
140

Lifespan Ogive
Installed 19721978: unknown & non-Interpace
Category 1
th
th
th
145
40
35
10,000
8,000
30
25
6,000
20
Category 2 & 3
th
th
th
15
10
4,000
2,000
5
0
0
10
15
20
25
30
35
40
~2010
~2015
45
50
Age (years)
~1975
~1980
~1985
~1990
~1995
~2000
~2005
Figure 5.8 Lifespan ogive installed 1972-1978: unknown and non-Interpace
~2020
~2025
12,000
45

Lifespan Ogive
Installed 19791991
40
Category 1
th
th
th
146
35
30
6,000
5,000
25
4,000
20
3,000
15
Category 2 & 3
th
th
th
10
2,000
1,000
0
0
10
15
20
25
30
35
40
~2020
~2025
45
50
Age (years)
~1985
~1990
~1995
~2000
~2005
~2010
~2015
~2030
~2035
7,000

Lifespan Ogive
Installed 19922007
1,400
1,200
1,000
0.8
Category 1
Sample pop. = 1 failure
th
th
th
0.6
800
600
0.4
Category 2 & 3
th
th
th
0.2
400
200
0
0
10
15
20
25
30
35
40
~2035
~2040
45
50
Age (years)
~2000
~2005
~2010
~2015
~2020
~2025
~2030
~2045
~2050
147
1.2
ReliaSoft W eibull++ 7 - www.ReliaSoft.com
Probability - Weibull: All Samples (category 1)
99.000
Probability-W eibull
90.000
all samples ( cat1)
W eibull-3P
RRX SRM MED FM
F=393/ S=4951634
Adj Points
U nadj Points
Adjusted Line
U nadjusted Line
50.000
10.000
5.000
148
Probability of failure (percent failed per sticks produced)
1.000
0.500
0.100
0.050
0.010
0.005
xx 4
3
3
2xx 4
2
x 2x 2
x2 x2
0.001
5.000E-4
xx 2
2
x2
xx 22
x22x2xx2233
x2xx4
2xx3
22
23
x x22 xx4
x
x
6
6
2
2
27
2
x2x 2x7
xx 13
xxx 13
2
2
x3
2x 3
2
x7
x 22
xx 7
xx 2
2
x2
1.000E-4
5.000E-5
1.000E-5
Brien Clark
Schiff Associates
8/ 1/ 2007
12:04:42 PM
0.1
1.0
10.0
Age (years)
=1.0585, =1.8339+5, =0.7750, =0.9801
Figure 5.11 Probability Weibull: all samples (Category 1)
100.0
Probability - Weibull: All Samples (category 1 & 2)
99.000
90.000
all samples ( cat1+2)
W eibull-3P
RRX SRM MED FM
F=758/ S=4951634
Adj Points
U nadj Points
Adjusted Line
U nadjusted Line
50.000
10.000
149
5.000
1.000
0.500
0.100
0.050
0.010
0.005
0.001
xx 2
2
5.000E-4
xx 2
2
x2
xx 2
2
x7
x 22
xx5
4x5
45x5x5x5x57
2x3
2x24x4x66
x4
2xx4
2xx3x33
xx2x2x3
2x3
222
2x5
2
x2x2x8
4
4
2x8
2x5
x6
xx9
x
x
9
3
3
2
2
x
x
6
5
5
4
4
3
3
2
2
x
x
8
8
5
5
4
4
37
366
xxx7
3
2x7
27
3
3
2x5
2223
xx6
3x6
3x5
2
2
x7
x8
7
7
2x 8
2
x13
x7
5
5
2
2
xx xx13
23
2
x3
2x 3
23
x6
3
3
2x 6
2
xx 4
3
3
2xx 4
2
xx 6
2 xx 6
2
x2 x2
x2
xx 2
2
1.000E-4
5.000E-5
1.000E-5
Brien Clark
Schiff Associates
8/ 1/ 2007
12:38:08 PM
0.1
1.0
10.0
Age (years)
=1.1602, =5.0682+4, =0.6975, =0.9832
Figure 5.12 Probability Weibull: all samples (Categories 1 and 2)
100.0
Probability - Weibull: All Samples (category 1, 2, & 3)
99.000
90.000
all samples ( cat1+2/ 3)
W eibull-3P
RRX SRM MED FM
F=25217/ S=4951634
Adj Points
U nadj Points
Adjusted Line
U nadjusted Line
50.000
10.000
150
5.000
1.000
x851
11
2
5xx7
x1325
266
x5
45x5 x165
70
xxxx835
1325
1071
1071
835
275
60
3
6
3
6
xxxx3696
4275
14
685
x
x
x
528
528
3
11
3
xxx 490
1628
1628
23
2
385
385
xxxxxx2348
x2348
10
10
4490
4
100
2100
4
2
4
xxxx2000
2000
x
x
x
x
2
3
2
3
2
2
x
x
370
370
xx 1322
1322
23
3
2
5
xxxx2785
2785
25
2
0.500
0.100
28
4
2
4
xx17
23
28
12
12
2x17
2x3
6
6
5
5
27
4
2
4
xx13
68
6xx7
7
7
5
5
213
3
4
2xx8
3
4
xxxx425
425
8
8
7
7
2278
3
2
3
x5
x1300
3
3
2x5
2xx2xx2278
xxx1300
0.050
xx8
xx 91
35
35
25
25
13
13
6
6
3
3
291
2
7
7
3
3
2
2
6
6
3
3
2x7
2x 8
13
13
3
3
2x14
2x7
2x5
2x14
4
4
3
3
27x
2xx5
4
4
3
3
26
2xx7x
27
2xx6
x x2 2 x x7
x 607
x 607
0.010
x 340
x 340
0.005
0.001
xx 3
2 xx 3
2
5.000E-4
xx 2
2
xx 3
2xx 3
2
xx 2
2
1.000E-4
5.000E-5
1.000E-5
Brien Clark
Schiff Associates
8/ 1/ 2007
12:39:36 PM
0.1
1.0
10.0
Age (years)
=2.0050, =535.2077, =0.3250, =0.9535
Figure 5.13 Probability Weibull: all samples (Categories 1, 2, and 3)
100.0
Probability - Weibull: All Samples (category 3)
99.000
90.000
all samples ( cat 3)

W eibull-3P
RRX SRM MED FM
F=24459/ S=4951634
Adj Points
U nadj Points
Adjusted Line
U nadjusted Line
50.000
10.000
151
5.000
1.000
xx851
11xx 1325
xx 851
11
266
266
70 x 165
70x 165
xxx1325
1071
x 835
1071
xx60
835
275
275
3
60
3
14
14
xx3696
xxxx3696
xx528
685
685
11
11
528
xx1628
xxxx1628
x10
490
490
385
385
10
x x2348
xxx2348
x
99
x
99
x 2000
x 2000
370
370
x x1322
xx1322
x 2785
x 2785
0.500
0.100
x 13 x 17
x 425
x 1299 x 278
0.050
x 607
0.010
x 13
x 340
xx 91
25
x 35
13
x 607
x 17
13
xx425
x 278
x 1299
x 91
25
13
x 13xx 35
x 340
0.005
0.001
Brien Clark
Schiff Associates
8/ 1/ 2007
1:02:11 PM
1.0
10.0
Age (years)
=4.8065, =154.8987, =18.0650, =0.9659
Figure 5.14 Probability Weibull: all samples (Category 3)
100.0
Probability - Weibull: Installed pre-1955 (category 1)
99.000
90.000
pre-1955 ( cat1)
W eibull-3P
RRX SRM MED FM
F=32/ S=476376
Adj Points
U nadj Points
Adjusted Line
U nadjusted Line
50.000
10.000
152
5.000
1.000
0.500
0.100
0.050
0.010
x2
0.005
x2
0.001
5.000E-4
1.000E-4
Brien Clark
Schiff Associates
8/ 1/ 2007
1:03:30 PM
1.0
10.0
Age (years)
=4.2785, =429.8055, =13.5775, =0.9931
Figure 5.15 Probability Weibull: installed pre-1955 (Category 1)
100.0
Probability - Weibull: Installed pre-1955 (category 1 & 2)
99.000
90.000
pre-1955 ( cat1+2)
W eibull-3P
RRX SRM MED FM
F=396/ S=476376
Adj Points
U nadj Points
Adjusted Line
U nadjusted Line
50.000
153
10.000
5.000
1.000
0.500
xx5
4x5
455x5x5 55
xx3
xxx3xx33
2
2x224x4x6x6
2x4
2
x5
xx2x2x2x222x4
2
2
4x8
4x5
x6
x9x9x2x2x8
x
x
x
x
6
5
5
4
4
x833
xx58
xxx7
5
4
4
37
366
xxxxxx7
3
3
27
2
xx3xx35
5
3
3
2
2
x
x
x7x88
xx xx7
55
x x2x2x3 322
xx 6
x
6
3
x3
2
2
xx 3
x3
x2
2xx 2
2
x 6x 6
x 2x 2
0.100
0.050
0.010
0.005
x2 x2
x2
x2
0.001
5.000E-4
1.000E-4
Brien Clark
Schiff Associates
8/ 1/ 2007
1:07:17 PM
0.1
1.0
10.0
Age (years)
=1.2624, =8407.7379, =0.5975, =0.9865
Figure 5.16 Probability Weibull: installed pre-1955 (Categories 1 and 2)
100.0
Probability - Weibull: Installed 1955-63 (category 1)
99.000
90.000
1955-63 ( cat1)
W eibull-3P
RRX SRM MED FM
F=40/ S=1049061
Adj Points
U nadj Points
Adjusted Line
U nadjusted Line
50.000
10.000
154
5.000
1.000
0.500
0.100
0.050
0.010
0.005
x 7 xx 27x 2
x2
x2
0.001
5.000E-4
1.000E-4
5.000E-5
1.000E-5
Brien Clark
Schiff Associates
8/ 1/ 2007
1:08:19 PM
1.0
10.0
Age (years)
=1.4728, =3.5017+4, =4.7700, =0.9829
Figure 5.17 Probability Weibull: installed 1955-63 (Category 1)
100.0
Probability - Weibull: Installed 1955-63 (category 1 & 2)
99.000
90.000
1955-63 ( cat1+2)
W eibull-3P
RRX SRM MED FM
F=73/ S=1049061
Adj Points
U nadj Points
Adjusted Line
U nadjusted Line
50.000
10.000
155
5.000
1.000
0.500
0.100
0.050
0.010
0.005
xx 2
2
x 2x 2
x 5x x7 5x 7
x 5xx45x 5x 5
x
4
x2 x2
xx 2
2
0.001
x2
5.000E-4
x2
1.000E-4
5.000E-5
1.000E-5
Brien Clark
Schiff Associates
8/ 1/ 2007
1:09:35 PM
1.0
10.0
Age (years)
=1.5723, =1.6802+4, =5.0325, =0.9896
Figure 5.18 Probability Weibull: installed 1955-1963 (Categories 1 and 2)
100.0
Probability - Weibull: Installed 1955-63 (category 1, 2, & 3)
99.000
90.000
1955-63 ( cat1+2/ 3)
W eibull-3P
RRX SRM MED FM
F=2422/ S=1049061
Adj Points
U nadj Points
Adjusted Line
U nadjusted Line
50.000
10.000
156
5.000
1.000
0.500
xx11
x112
x2851
xx851
x 165
55xxx 7165
5x 7
4x 5x
5x
2 x x1322
3
2 xx34xxx35x
3
x x1322
0.100
0.050
0.010
0.005
0.001
x3
5.000E-4
x3
1.000E-4
5.000E-5
1.000E-5
Brien Clark
Schiff Associates
8/ 1/ 2007
1:10:18 PM
1.0
10.0
Age (years)
=4.1361, =158.2652, =3.3175, =0.8605
Figure 5.19 Probability Weibull: installed 1955-63 (Categories 1, 2, and 3)
100.0
99.000
1955-63 ( cat3)
W eibull-3P
RRX SRM MED FM
F=2349/ S=1049061
Adj Points
U nadj Points
Adjusted Line
U nadjusted Line
90.000
157
50.000
10.000
5.000
1.000
0.500
11
xx 851
x 1322
11
x 851
x 165
0.100
10.0
Brien Clark
Schiff Associates
8/ 1/ 2007
1:11:21 PM
x 165
x 1322
100.0
1000.0
Age (years)
=51.8054, =2528.5392, =2198.8000, =0.8272
10000.0
99.000
90.000
1964-67 ( cat 1)
W eibull-3P
RRX SRM MED FM
F=31/ S=594178
Adj Points
U nadj Points
Adjusted Line
U nadjusted Line
50.000
10.000
158
5.000
1.000
0.500
0.100
0.050
0.010
0.005
x2
x7
x2
x7
0.001
5.000E-4
1.000E-4
Brien Clark
Schiff Associates
8/ 1/ 2007
1:12:05 PM
0.1
1.0
10.0
Age (years)
=0.9444, =6.5262+5, =4.4975, =0.9707
100.0
99.000
90.000
1964-67 ( cat 1+2)

W eibull-3P
RRX SRM MED FM
F=81/ S=594178
Adj Points
U nadj Points
Adjusted Line
U nadjusted Line
50.000
159
10.000
5.000
1.000
0.500
0.100
0.050
0.010
x x2 2 xx22
x7
x7
0.005
x6
x 5x 5
x5
xxx 5
4xx5
45
x6
0.001
5.000E-4
1.000E-4
Brien Clark
Schiff Associates
8/ 1/ 2007
1:13:14 PM
1.0E-2
0.1
1.0
Age (years)
=0.8886, =6.1712+5, =3.9000, =0.9643
10.0
100.0
99.000
90.000
1968-71 ( cat 1)
W eibull-3P
RRX SRM MED FM
F=60/ S=551218
Adj Points
U nadj Points
Adjusted Line
U nadjusted Line
50.000
10.000
160
5.000
1.000
0.500
0.100
0.050
0.010
x 13
0.005
x3
x 13
x3
0.001
5.000E-4
1.000E-4
Brien Clark
Schiff Associates
8/ 1/ 2007
1:14:07 PM
0.1
1.0
10.0
Age (years)
=1.3053, =1.6093+4, =2.5150, =0.9429
100.0
99.000
90.000
1968-71 ( cat 1+2)

W eibull-3P
RRX SRM MED FM
F=82/ S=551218
Adj Points
U nadj Points
Adjusted Line
U nadjusted Line
50.000
161
10.000
5.000
1.000
0.500
0.100
0.050
x3
0.010
xx 3
x2
2
0.005
x2
x2
x2
x3
xx 3
x2
2
x2
0.001
5.000E-4
1.000E-4
Brien Clark
Schiff Associates
8/ 1/ 2007
1:15:37 PM
0.1
1.0
10.0
Age (years)
=1.1414, =3.6250+4, =2.7550, =0.9540
100.0
99.000
90.000
1972-78 ( cat 1)
W eibull-3P
RRX SRM MED FM
F=194/ S=838154
Adj Points
U nadj Points
Adjusted Line
U nadjusted Line
50.000
10.000
162
5.000
1.000
0.500
0.100
0.050
0.010
xx 4
3
3
2xx 4
2
x 2x 2
x2 x2
0.005
x2
x2
0.001
5.000E-4
x2
26
2
x xx2x26
x x2 2
3
2x3
22x2x 33
xxxx4x4
x2
x2
x2
1.000E-4
5.000E-5
1.000E-5
1.0E-2
Brien Clark
Schiff Associates
8/ 1/ 2007
1:16:23 PM
0.1
1.0
Age (years)
=1.0019, =9.5582+4, =0.9150, =0.9704
10.0
100.0
99.000
90.000
1972-78 ( cat 1+2)
W eibull-3P
RRX SRM MED FM
F=350/ S=838154
Adj Points
U nadj Points
Adjusted Line
U nadjusted Line
50.000
10.000
163
5.000
1.000
0.500
0.100
0.050
0.010
0.005
x2
x2
x2
x3
6x 3
6
xx 4
3
2 xxx 4
2
x3
x2 x2
x2 x2
x2
2x4
2x33x3x 66
2x3
2x4
2xxx3
3
3
2x9
22x244
x6xx66
4x6
49
x7
7
4
4
3x7
3 33
x3xx7
3
3
x
x
3
x
x
6
6
xx 5x2 22 2
x3
2xx 3
25
x2
x2
0.001
x2
5.000E-4
x2
1.000E-4
5.000E-5
1.000E-5
Brien Clark
Schiff Associates
8/ 1/ 2007
1:18:24 PM
0.1
1.0
10.0
Age (time)
=1.1349, =2.2255+4, =0.8500, =0.9790
100.0
99.000
90.000
1972-78 ( cat 1+2/ 3)
W eibull-3P
RRX SRM MED FM
F=15354/ S=838154
Adj Points
U nadj Points
Adjusted Line
U nadjusted Line
50.000
10.000
164
5.000
266
266
70
70
xxxxx1325
1325
1071
1071
xx3696
835
275
275
3835
6
3
6
60
60
14
14
xxxx3696
xx528
685
685
xxxxx1628
1628
3528
3
xxx490
490
385
385
4
4
xx2000
10
100
4
4
2x100
210
xxx2000
1.000
0.500
x370
3
xxx370
23
22
2
2x12
2x244
6
4x7
412
x425
7
3
4
3
4
2x7
26x67
x x425
x x278
278
38
3
x3xx32x28
x91
xx xxx91
35
35
25
25
x6
26
2
xx 6x3x 3
2xx 3
26
7x 3
4
2x 4
2
4
4
3
3
2 xx 6
27
2 xx22 xx 26
x 340 xx340
0.100
0.050
0.010
0.005
x3
x3
x2
0.001
x2
5.000E-4
x3
x3
x2
x2
1.000E-4
5.000E-5
1.000E-5
Brien Clark
Schiff Associates
8/ 1/ 2007
1:19:02 PM
0.100
1.000
10.000
Age (years)
=1.6758, =667.3259, =0.8450, =0.9303
100.000
99.000
1972-78 ( cat 3)
W eibull-3P
RRX SRM MED FM
F=15004/ S=838154
Adj Points
U nadj Points
Adjusted Line
U nadjusted Line
90.000
165
50.000
10.000
5.000
266
70
xx1325
x 1071
x 835
x
275
60
3
14
xx3696
1.000
x266
70
x x1325
x 1071
x 835
x275
x60
3
14
x x3696
x 685
528
xx1628
x 490
x99
385
10
xxx2000
0.500
x 685
528
x x1628
x 490
x x385
10
99
x 2000
x 370
x 370
x 425
x 425
x 278
0.100
0.050
0.010
x 278
x 91
xx 35
25
x 340
x 91
x 35
25
x x340
Brien Clark
Schiff Associates
8/ 1/ 2007
1:21:19 PM
1.0
10.0
100.0
Age (years)
=118.5320, =810.5559, =749.8400, =0.9661
1000.0
99.000
90.000
1979-91 ( cat 1)
W eibull-3P
RRX SRM MED FM
F=35/ S=1061650
Adj Points
U nadj Points
Adjusted Line
U nadjusted Line
50.000
10.000
166
5.000
1.000
0.500
0.100
0.050
0.010
0.005
x2
x2
x 2x 2
xx22
0.001
5.000E-4
1.000E-4
5.000E-5
1.000E-5
Brien Clark
Schiff Associates
8/ 1/ 2007
1:22:30 PM
0.1
1.0
10.0
Age (years)
=0.5358, =4.5971+9, =0.8425, =0.9398
100.0
99.000
90.000
1979-91 ( cat 1+2)
W eibull-3P
RRX SRM MED FM
F=116/ S=1061650
Adj Points
U nadj Points
Adjusted Line
U nadjusted Line
50.000
10.000
167
5.000
1.000
0.500
0.100
0.050
x22
x x8x2x8x2x2
2x5x825
2
x5xx8x5
x
3
3
x 7 xx 7
x5
3
28xxx25
2 2
x 8 xx3
x7 x7
x2 x2
x2 x2
0.010
0.005
x2
0.001
x2
5.000E-4
1.000E-4
5.000E-5
1.000E-5
Brien Clark
Schiff Associates
8/ 1/ 2007
1:23:40 PM
0.1
1.0
10.0
Age (years)
=1.5554, =8595.6789, =2.4600, =0.9335
100.0
99.000
90.000
1979-91 ( cat 1+2/ 3)
W eibull-3P
RRX SRM MED FM
F=5899/ S=1061650
Adj Points
U nadj Points
Adjusted Line
U nadjusted Line
50.000
10.000
168
5.000
xx 2348
2348
1.000
xxx222
xx2785
2785
0.500
0.100
x 607 x 607
0.050
x3
xx17
3
2
288
2xx5
217
3xx5
38xx85
2xx7
25
x13
5x
3
3
25x
2xxx7
xx13
x7
2x 7
2x8x 8
xx 14
x 2xx214
0.010
0.005
x2
x2
0.001
x2
x2
5.000E-4
1.000E-4
5.000E-5
1.000E-5
Brien Clark
Schiff Associates
8/ 1/ 2007
1:24:31 PM
0.1
1.0
10.0
Age (years)
=1.3825, =2214.7654, =0.7525, =0.9357
100.0
99.000
90.000
1992-2006 ( cat 1)
W eibull-2P
MLE SRM MED FM
F=1/ S=380997
D ata Points
Probability Line
50.000
10.000
169
5.000
1.000
0.500
0.100
0.050
0.010
0.005
0.001
5.000E-4
1.000E-4
Brien Clark
Schiff Associates
8/ 1/ 2007
1:25:21 PM
10.0
100.0
Age (years)
=6.9731, =76.8097
99.000
90.000
1992-2006 ( cat 3)
W eibull-2P
MLE SRM MED FM
F=1299/ S=380997
D ata Points
Probability Line
50.000
10.000
170
5.000
1.000
0.500
0.100
0.050
0.010
0.005
0.001
5.000E-4
1.000E-4
Brien Clark
Schiff Associates
8/ 1/ 2007
1:25:56 PM
10.0
100.0
Age (years)
=6.9861, =27.4499

Weibull-Predicted 10-Year Failure Probabilities by Time Period
1.00E-02
2.50E-02
171
Probability of a category 1 failure (percent)
8.00E-03
2.00E-02
7.00E-03
6.00E-03
1.50E-02
5.00E-03
4.00E-03
1.00E-02
3.00E-03
2.00E-03
5.00E-03
1.00E-03
0.00E+00
0.00E+00
All
pre-1955
195563
196467
196871
197278
1942-2007
all
Figure 5.34 Weibull-predicted 10-year failure probabilities by time period
197991
19922006
Probability of a Category 1 or 2 failure (percent)
9.00E-03
0.4
0.4
0.35
0.35
0.3
0.3
0.25
0.25
0.2
0.2
0.15
0.15
0.1
0.1
0.05
0.05
0
All
pre-1955
195563
196467
196871
197278
1942-2007
all
Figure 5.35 Weibull-predicted 100-year failure probabilities by time period
197991
19922006
Probability of a category 1 or 2 failure (percent)
172
Probability of a category 1 failure (percent)

Weibull-Predicted 100-Year Failure Probabilities by Time Period
CHAPTER 6
CONCLUSIONS
The research objectives of this AwwaRF project were twofold: 1) development of a general
PCCP evaluation matrix, and 2) statistical analysis of PCCP failure data.
A general evaluation matrix to assist water utilities identify PCCP at risk of failure in their systems
began with research prepared for the proposal for this project. The matrices developed for both LC-type
PCCP and EC-type PCCP provide a tested tool to PCCP-owning utilities. The tool is intended to be
utilized as a screening method to allow allocation of resources toward the PCCP at greatest risk of failure
within their systems. The resultant matrix is included as Appendix A for this purpose.
The statistical analyses of PCCP failure data took an actuarial or insurance view of the
population of PCCP. Based on data made available to the investigators, there was a population
of about 100 million feet of PCCP pipe produced between WWII and 2006 in the United States.
The failure data were characterized based on three forms of failure: catastrophic ruptures,
failure discerned by inspection, and loss of service which were classified as Category 1,
Category 2, and Category 3 failures, respectively. Then the data were grouped based upon the
AWWA standard in effect in the year of manufacture. Histograms were prepared that allow both
determination of failure frequency by year of manufacture as well as comparison of the relative
differences in failure frequency. Statistical analyses of the populations using ogive and Weibull
methods allowed preliminary estimates of remaining service life of the PCCP pipelines. The
complete failure database (Table 1.1) is provided for independent review and for any future
analysis the reader might wish to undertake. Histograms, ogives, bar graphs, and Weibull
probability failures are included to assist the PCCP utility owner and are included as Figures 3.1
through 3.20 and Figures 5.1 through 5.35.
From the first issue of PCCP standard 7B.2-T (1949), the AWWA PCCP standards were
revised in 1952, 1955, 1958, 1964, 1972, 1979, 1984, 1992, 1999, and 2007. Each of the
standards revisions were studied, and 57 items within those which were revised at least once
were tabulated and compared, including:
Size and pressure ranges for EC and LCP

Quality of concrete and mortar
Materials and dimensions of steel cylinders
Materials and dimensions and allowable stresses of wire reinforcement
Core tolerances
Joint tolerances
Coating thickness and quality
Some notable trends were discerned from the material changes in the standards including:
Initial conservatism in Standard 7.2-T (1949) was increased for some parameters in
AWWA C301-52.
Subsequent revisions to the standard prior to 1984 generally resulted in reduced
dimensional minimums and higher effective component stresses until the 1984
revision. The net effect was the highest rate of failure for pipe manufactured between
1967 and 1979. Fully 50 percent of the catastrophic ruptures and leaks recorded were
173
of pipe manufactured between 1971 and 1979. See Figure 6.1. Most of the pipe
inspected and found to be defective (61 percent) were made within the same time
frame. See Figure 6.2.
Significantly increased product requirements were included in AWWA C301-92, and
significantly increased design basis complexity was added in AWWA C304-92. The
PCCP made to that standard have performed statistically significantly better than
their predecessors.
The reasons for PCCP failures are many. They included:
Rupture or break broken wires found after the failure many causes
Leaking at joints many causes including out-of-roundness of joint and construction
damage
Cracks in core many causes including alkali-silica reactivity of the aggregate
Low quality of core poor concrete strength
H2S (force mains) unlined
Dented cylinder fabrication and construction
Cracks in cylinder welds poor cylinder fit-up
Low quality of wires not just Type IV
Overwrapping of wire inadequate total prestress
Wire spliced and re-stressed inadequate total prestress
Low quality of mortar low density, low thickness, and low cement content
High chlorides in soil corrosive/aggressive soil inappropriate for mortar-coated pipe
Inadequate joint restraint pipe moved exposing joint to environment
Construction damage coatings damaged and not repaired
Coating delamination many causes
Hydrogen embrittlement of wire excessive cathodic protection applied to
susceptible wire
Inadequate prestress wires broken and spliced without re-tensioning resulting in
low core compression
Cantilever (bending or broken back) many causes including poor bedding
Settlement general and at structures
Poor bedding not corresponding to design assumption
Surge unanticipated and above design value
Looped gasket joint fit-up
Wrong pipe class pipe laid out of order
Cracks in joint welds poor/no field inspection
Hydro test pressure in excess of design pressure
Excess external load greater than design assumption
Missing joint coating
This variety of causes is related to the complexity of the product. It requires a good deal
of attention during the design, manufacture, installation, and operation in order to be successful.
Failure to pay appropriate attention to any part of the design, manufacturing, inspection,
construction, and operation of PCCP can lead to failure of the pipeline.
174
EXPECTED SERVICE LIFE FOR PCCP
As anything gets older, it gets closer to the end of its useful life. The same is true for
pipe, its just that the life expectancy varies by many factors. Life expectancy varies from 50
years, to 100 years, to indefinite depending on the perception of the pipeline owner.
Some of the utilitys options to assess the service life beyond the statistical methods
presented in this report include:
1.
2.
3.
4.
5.
6.
Develop a risk-based assessment protocol to determine from that basis which of

the utilitys PCCP is of the greatest risk for failure. Development of a risk-based
assessment is addressed in previous studies (Romer and Bell 2004; Kleiner,
Sadiq, and Rajani 2004 and 2005).
Reduce operating pressures below cracking stress of the core restrained by the
steel cylinder only. If that is not feasible, reduce operating pressures below yield
strength of the steel cylinders. Both of these require structural analysis because
they are not part of the original design basis. These are short-term actions that
may result in a longer time available to investigate and evaluate alternative
courses of action.
Review the utilitys PCCP pipeline operating practices to make sure that rapid
variations in pressure, such as due to variable flow rate and valve closure do not
occur.
Conduct nondestructive evaluations. Those evaluations can include: (a) internal
inspection looking for cracks and delamination; (b) internal inspection of the pipe
by indirect assessment methods; and (c) acoustic monitoring to determine if wires
are actively breaking.
Conduct an external inspection of the pipe, by excavation and visual inspection of
pipe. At that time, consider removal of coating for petrographic testing of the
mortar for soundness and chemical constituents and removal of the coating in the
identified area(s) and inspect the wire. The pipeline should be depressurized prior
to excavation. Consideration should also be given to having a rehabilitation or
replacement plan in-place prior to excavation.
Should the investigations and risk assessment, coupled with the corrosion and
other assessments result in the need to remediate the pipeline, consideration can
be given to (a) structural rehabilitation of selected segments of the pipeline;
(b) removal and replacement of selected segments of the pipeline; or
(c) replacement of the pipeline.
Each of these options should be evaluated in detail prior to execution. The determination
of which option or options may have priority should be aided by a risk-based evaluation (Item 1
above), which is all beyond the scope of this study.
175
197991
197278
9%
Unknown & nonInterpace
11%
19922007
0%
pre-1955
8%
195563
10%
196467
8%
196871
15%
197278
Interpace
39%
Figure 6.1 Category 1 PCCP failures by manufacturing era

19922007
5%
pre-1955
0%
195563 196467
10%
0%
196871
0%
197991
24%
197278
Interpace
18%
197278
Unknown & nonInterpace
43%
Figure 6.2 PCCP failures by manufacturing era
176
CHAPTER 7
FUTURE STUDY RECOMMENDATIONS
Future study recommendations:
Determine effectiveness of NDT examinations of PCCP in reducing the rate and

frequency of PCCP failures. The investigators could only determine that the most
probable cause of the failure rate reduction in pipe manufactured and installed since
1992 is the increase in conservatism in design and manufacturing since the 1992
standard was adopted.
Complete the production database by obtaining production data from manufacturers
unwilling to cooperate with this study, then revise the statistical analyses.
Expand the failure database. Utilities not included as participants in this study
invariably know of previous failures not listed here. The larger the sample database,
the better the statistical analysis and predictive modeling will be. A survey or
information request kit may be distributed to collect additional database entries.
Maintain the failure database. This will require annual funding and the cooperation
of the AwwaRF subscribers to contribute data that may otherwise be kept private
pending possible litigation. It will be particularly interesting to gain a better insight
on performance of LC-PCCP.
Study the effects of uncontrolled surge on the performance of PCCP. Case studies
appear to be the most appropriate approach.
Study the limits of soil chemistry and resistivity that have been recommended for
PCCP longevity and determine their effectiveness. That would include a program of
field soil tests and physical examination of successful PCCP pipelines.
Integrate the statistics developed in this study with a risk management program. A
case study or two may be appropriate. The mathematics developed for managing the
risk inherent with PCCP pipes must not be daunting to management.
Determine the cost effectiveness of repair techniques in extending the service life of
PCCP.
177
APPENDIX A:
QUALITATIVE ASSESSMENT MATRIX
179
Appendix A1: Short Form

Assessment of Prestressed Concrete Cylinder Pipe
Pipeline:
Diameter (in) :
Pressure Range (psi):
No.
Score
DON'T
NO
KNOW
YES
(score -1) (score 0) (score +1)
Factor
1 Was the pipeline constructed prior to 1964 or after 1992?

Is the pipe manufacturer still in business on the site at which
2 the pipe was made?
Is the pipe constructed with 6 gauge (0.192" diameter) or larger
wire? (Score this -1 if the pipe constructed with 8 gauge
3 (0.162" diameter) wire.)
Was inspection done at the manufacturing site and at the time
4 of installation by the owner or the owner's consulting engineer?
Is surge unlikely on the pipeline? (Score -1 if pumped flow or
5 other condition where surge is likely)
Subtotals
Total Assessment Score
Brief description of pipeline:
180
0
0
Appendix A2: Long Form

Failure Assessment of Prestressed Concrete Cylinder Pipe
Pipeline:
Diameter (in) :
No.
Score
DON'T
NO
KNOW
YES
(score -1) (score 0) (score +1)
Question
Factors which may indicate an appropriate DESIGN basis:

Did the project specification include requirements in addition to
1 AWWA Standard C301 or the USBR std. spec?
Did the Project Contract Documents include minimum design
requirements (wire size and spacing, etc.?) (Score this -1 if the
2 pipe manufacturer submitted the design.)
Was the design bedding angle less than or equal to 90 degrees?
(score this -1 if 120 degree bedding angle or greater was
3 assumed.)
Factors which may indicate an appropriate
MANUFACTURING quality:
4 Is the pipe manufacturer still in business?
5 Was the pipeline constructed prior to 1964 or after 1992?
Was the pipe manufactured in a fixed plant? (Score this -1 if the
6 pipe was manufactured on-site)
Is the prestressing wire Class I or Class II? (Score this -1 if Class
7 III or Class IV.)
Is all the reinforcing wire > No. 6? (Score this -1 if No. 8 wire
8 was used.)
Is the steel cylinder 16 ga. or thicker? (Score this -1 if 18 ga.
9 cylinder is used.)
Is the design wire stress ratio < 70%? (Score this -1 if 75% of
10 ultimate strength.)
11 Is the coating, cast or mortar, at least 7/8 inch thick?
Was mortar slurry placed under the prestressing wires? (This was
12 required after 1984.)
Were bonding straps placed under wires to allow corrosion
monitoring? (Score this +1 if shorting cables were installed
13 between pipe segments.)
Factors which may indicate appropriate INSPECTION was
done:
181

Pipeline:
Diameter (in) :
No.
Score
DON'T
NO
KNOW
(score -1) (score 0)
Question
14 Were pipe design submittals provided for review by owner?

Was documentation of inspection & testing by manufacturer submitted to
15 the owner for review?
16 Was any manufacturing inspection done by the owner or owner's engineer?
Was any on-site construction inspection done by the owner or owner's
17 engineer?
Were wire tests done in accordance with the then-current standards (i.e. mill
18 certificates) and submitted to the owner for review?
Did the pipe manufacturer provide on-site assistance to the installation
19 contractor?
Factors which may indicate that appropriate care was taken

during CONSTRUCTION:
Was the pipe placed on shaped bedding? (90 deg. shaped bedding is the
20 typical design assumption.)
Was imported material utilized for bedding? (Score this -1 if native soils
21 were used for bedding.)
Was the hydrostatic test passed on the first attempt? (May otherwise be
22 indicative of rolled gaskets.)
Was the hydrostatic test pressure no more than 20% over the design
23 operating pressure?
24 Were joint bonds installed and tested?
(If yes to the above question) Is the pipeline electrically isolated from other
25 pipelines?
Were joints welded for thrust restraint? (Score this -1 if bolted joint
26 restraints, which allow slight movement, were used.)
27 Were joints mortared inside and out?
Factors which may indicate appropriate OPERATION and

MAINTENANCE:
Is surge unlikely on the pipeline? (Score -1 if pumped flow or other condition
28 where surge is likely)
182
YES
(score
+1)

Pipeline:
Diameter (in) :
No.
Score
DON'T
NO
KNOW
(score -1) (score 0)
Question
YES
(score
+1)
Is the earth fill over the pipeline substantially unchanged from when it was
29 constructed?
30 Is the pipeline subject to regular corrosion monitoring?

If cathodic protection is applied, is the voltage applied < 850 mv? (Score this
31 -1 if > 1000 mv and score it 0 if between these values or no C-P.)
Other Factors which may indicate CONDITION:

32
33
34
35
The pipeline has experienced neither a leak nor a break. (Score this -1 if it
has leaked, etc.)
If an internal inspection has been done (of any type) there is little or no
indication of damage/cracks/spalls/broken wires. (Score this -1 if
damage/cracks/spalls/broken wires are indicated.)
If soil corrosivity tests were performed were all laboratory saturated soil
esistivities greater than 1000 ohm-cm or field resistivities greater than 1500
ohm-cm found at pipe depth?
If soil chemistry tests were performed were all chloride levels less than 350
ppm (mg/kg)?
36 If soil chemistry tests were performed were pH values greater than 5.5 ?
If close interval pipe to soil potentials have been measured, are any areas
greater than 50 feet in length more negative than -300 millivolts to copper37 copper sulfate?
Groundwater levels are stable. (Score this -1 if any portion of the pipeline is
38 in an area of fluctuating groundwater levels.)
Subtotals
Total Assessment Score
Brief description of pipeline:
183
0
0
GLOSSARY
Accelerator:
A chemical substance that increases the rate at which a chemical

reaction (e.g., curing) would otherwise occur.
Active:
(1) The negative direction of electrode potential.

(2) A state of a metal that is corroding without significant influence of
reaction product.
Alkyd:
Type of resin formed by the reaction of polyhydric alcohols and

polybasic acids, part of which is derived from saturated or unsaturated
oils or fats.
Anerobic:
Free of air or uncombined oxygen.
Anion:
A negatively charged ion that migrates through the electrolyte toward

the anode under the influence of a potential gradient.
Austenitic Steel:
A steel whose microstructure at room temperature consists

predominantly of austenite.
Anode:
(1) Corrosion: The electrode of a corrosion cell that has the greater
tendency to corrode or oxidize.
(2) Cathodic Protection: The expendable materials, which is buried
and through which direct current flows into the soil. Common
materials used for this purpose are graphite, high silicon iron,
magnesium, zinc, and scrap iron.
Anode Field:
The area in which the soil potential is raised because of current flow
away from a ground electrode (anode). The extent of this influence is a
function of soil resistivity and the magnitude of current flow.
Anodic Interference:
Current discharge from a structure caused by current pickup in an

anodic area. The current pickup results from a raised soil potential due
to the concentration of current flowing in the soil away from an anode.
Not always a problem, can be soil attenuation effect only.
Backfill:
The material (such as sand) which is placed around a buried pipeline.

Also, the material which is placed around anodes to ensure uniform
current discharge and to extend the useful life of the anode. Coal coke
or petroleum coke is used in conjunction with carbon or duriron
anodes, while a mixture of gypsum, bentonite and salt is used with
zinc, magnesium, and scrap iron anodes.
185
Bituminous Coating:
An asphalt or coal-tar compound used to provide a protective coating

for a surface.
Cathode:
The electrode of a corrosion cell where a net reduction reaction occurs.

In corrosion processes, the cathode is usually that area which does not
corrode.
Cathode Field:
The area in which the structure to soil potential is lowered because of

current flow. The extent of this influence is a function of soil
resistivity and current density in the soil.
Cathodic
Disbondment:
The destruction of adhesion between a coating and the coated surface

caused by products of a cathodic reaction, typically hydrogen.
Cathodic Interference:
Current discharge from one structure caused by an adjacent soil

potential which results from a concentration of protection current
flowing to an adjacent bare or poorly coated structure.
Cathodic Polarization:
That portion of the polarization of a cell which occurs at the cathode.
Cathodic Protection:
Reduction or prevention of corrosion of a metal surface by making it

cathodic, for example, by the use of sacrificial anodes or impressed
current rectifier protection systems. In cathodic protection, the
structure is made a part of an external electrical circuit in which
sufficient direct current flows through the surrounding electrolyte,
from an external anode, into the surrounding electrolyte, from an
external anode, into the structure. This current opposes the corrosion
cell currents discharged at the anodic (-) areas and the entire surface of
the structure is changed to a cathodic (+) or protected state; hence the
name cathodic protection.
Cation:
A positively charged ion that migrates through the electrolyte toward

the cathode under the influence of a potential gradient.
Cell:
A circuit consisting of an anode and a cathode in electrical contact in a

solid or liquid electrolyte. Corrosion generally occurs only at anodic
areas.
Examples Concentration Cell: A corrosion cell due to the potential
difference between the anode and cathode caused by differences in
composition of electrolyte.
Local Cell: A galvanic cell caused by small differences in composition
in the metal or the electrolyte.
Oxygen Concentration Cell: A galvanic cell caused by a difference in
oxygen concentration at two points on a metal surface.
186
Coating:
A liquid, liquefiable, or mastic composition that, after application to a

surface, is converted into a solid protective, decorative, or functional
adherent film.
Coating Resistance:
The electrical resistance of a coating to the flow of current. Unit of

measurement is ohms per square foot. Values range from 1,000 to
more than 1,000,000 ohms per square foot for conventional organic
coatings.
Concentration Cell:
An electrochemical cell, the electromotive force of which is caused by

a difference in concentration of some component in the electrolyte.
(This difference leads to the formation of discrete cathodic and anodic
regions.)
Contact:
A term used to describe an undesired metal-to-metal junction between

two different structures underground or defective and bypassed meter
or flange insulation.
Copper- Copper
Sulfate Electrode:
A standard or reference electrode used for determining potentials of

metals in soils or other electrolytes. It is the reference half-cell used
most in the field of corrosion. The half-cell, as it is known, is called a
saturated cooper-copper sulfate reference electrode. This is written as
such, CuCuSO4. The first Cu stands for the copper core which is
centered in the phenolic holder. The CuSO4 represents the copper
sulfate solution that is used to provide contact between the porous end
plug and the copper core. The term half-cell is used because the
electrode is only half of what is required to have a readable circuit.
With proper instrumentation, this half-cell can be used for many
measurements. The most common being the structure to soil potential.
This difference is created by the current flow through the electrolyte
(soil, water, etc.) around the structure and the resistance between the
point of reference and the structure.
If the structure is well coated, the half-cell is normally placed directly
over the structure because most of the resistance between the structure
and earth consists of the resistance of the coating itself. With a bare or
poorly coated structure, the resistance between structure and earth
consists of the contact resistance of the earth mass surrounding the
structure. This means that to read the maximum potential between the
structure and the reference it becomes necessary to place the half-cell
far enough away to compensate for the major part of this resistance.
187
When the half-cell is remote and the degree of cathodic protection is

being measured, both the on (current being drained) and off potentials
should be obtained. This will then indicate the degree of polarization
or holding potential for the subject structure.
Other uses will not be recapped; however, their use with the
Vibroground and other meters for stray current surveys may be
required at a later date.
Corrosion:
The deterioration of a material or its properties, usually a metal,

resulting from a reaction with its environment.
Corrosion Inhibitor:
A chemical substance or combination of substances that, when present

in the environment, prevents or reduces corrosion.
Corrosion Rate:
The rate at which corrosion damage proceeds.
Corrosion Potential
(Ecorr):
The potential of a corroding surface in an electrolyte relative to a

reference electrode under open-circuit conditions (also known as rest
potential, open-circuit potential, freely corroding potential, or freecorrosion potential).
Corrosion Resistance:
Property of a material, usually a metal, to withstand corrosion in a

given environment.
Corrosivity:
The tendency of an environment to cause corrosion.
Crevice Corrosion:
Localized corrosion of a metal surface at, or immediately adjacent to,

an area that is shielded from full exposure to the environment because
of close proximity of the metal to the surface of another material.
Critical Pitting
Potential (Ep, Epp):
The lowest value of oxidizing potential (voltage) at which pits

nucleate and grow. The value depends on the test method used.
Curing:
Chemical process of developing the intended properties of a coating or

other material (e.g., resin) over a period of time.
Current Density:
The current per unit area of metallic surface, usually expressed in

terms of milliamperes per square foot.
Deep Grounded:
One or more anodes installed vertically at a nominal depth of

15 meters (50 feet) or more below the earths surface in a drilled hole
for the purpose of supplying cathodic protection.
Depolarization:
The removal of factor resisting the current in an electrochemical cell.
Dielectric Coating:
A coating that does not conduct electricity.
188
Disbondment:
The loss of adhesion between a coating and the substrate.
Drain-Drainage:
Conduction of current from an underground metallic structure by

means of a metallic conductor to another underground structure,
usually by deliberate design, and/or conduction of current from an
underground metallic structure to a cathodic protection station.
Electrical Isolation:
The condition of being electrically separated from other metallic

structures or the environment.
Electro-Negative:
A term used to designate the metal most likely to corrode in a

bimetallic corrosion cell. Magnesium and zinc, for instance, are
electro-negative with respect to copper or steel. In the same
relationship, copper and steel are electro-positive.
Electrochemical Cell:
A system consisting of an anode and a cathode immersed in an

electrolyte so as to create an electrical circuit. The anode and cathode
may be different metals or dissimilar areas on the same metal surface.
Electrolysis:
The production of a chemical change in an electrolyte resulting from

the passage of electricity and often traditionally used to describe any
and all forms of corrosion.
Electrolyte:
An ionized chemical substance or mixture that will conduct electric

current, such as water, soil, and many chemical solutions.
Electrode:
Either the corroding or noncorroding portion of a corrosion cell; see

anode or cathode, whichever is appropriate. Also, used loosely to
describe half cells such as the copper-copper sulfate, silver-silver
chloride, and the calomel reference electrodes.
Embrittlement:
Loss of ductility of a material resulting from a chemical or physical

change.
Enamel:
(1) A paint that dries to a hard, glossy surface.

(2) A coating that is characterized by an ability to form a smooth,
durable film.
Epoxy:
Type of resin formed by the reaction of aliphatic or aromatic polyols

(like bisphenol) with epichlorohydrin and characterized by the
presence of reactive oxirane end groups.
Ferrite:
The body-centered cubic crystalline phase of iron-based alloys.
Film:
A thin, not necessarily visible layer of material.
189
Galvanic Cell:
A cell in which chemical change activates the source of electric

current. It usually consists of two dissimilar metals in contact with one
another, in a uniform or varying electrolyte, or two similar metals in a
dissimilar electrolyte.
Galvanic Series:
A list of metals and alloys arranged according to their relative

potentials in a given environment.
Graphitic Corrosion:
Deterioration of cast iron alloys in which the metallic constituents are

selectively leached or converted to corrosion products, leaving the
graphite intact.
Graphitization:
The formation of graphite in iron or steel, usually from decomposition

of iron carbide at elevated temperatures.
Grit:
Small particles of hard material (e.g., iron, steel, or mineral) with

irregular shapes that are commonly used as an abrasive in abrasive
blast cleaning.
Grit Blasting:
Abrasive blast cleaning using grit as the abrasive.
Groundbed:
One or more anodes installed below the earths surface for the purpose
of supplying cathodic protection.
Half-Cell:
A pure metal in contact with a solution of known concentration of its

own ion, at a specific temperature, develops a potential that is
characteristic and reproducible; when coupled with another half-cell,
an overall potential that is the sum of both half-cells develops.
Holidays:
A break or imperfection in a coating exposing the substrate or base

metal.
Hydrogen
Embrittlement:
A loss of ductility of a metal resulting from absorption of hydrogen.
Hydrogen Stress
Cracking:
Cracking that results from the presence of hydrogen in a metal in

combination with tensile stress. It occurs most frequently with highstrength alloys.
Impressed Current:
Cathodic protection current provided by rectifier-type protective

systems.
Inclusion:
A nonmetallic phase such as an oxide, sulfide, or silicate particle in a

metal.
Inorganic Zinc-Rich
Coating:
Coating containing a metallic zinc pigment (typically 75 wt% zinc or

more in the dry film) in an inorganic vehicle.
190
Insulation:
Fittings (e.g., unions, coupling, flanges) which permit metallic

separation of one section of structure from another without disturbing
the mechanical continuity and/or material used to provide physical
separation between external surfaces of structures.
Ion:
An ion is an atom or molecule normally electrically neutral that gains

or loses one or more electrons or protons.
Long Line Current:
Current flowing through the earth from an anodic to a cathodic area

and then returning through an underground metallic structure. Usually
found in areas where structures are continuous for a considerable
distance and where the current results from galvanic cell or foreign
action.
Mill Scale:
The oxide layer formed during hot fabrication or heat treatment of

metals.
Noble:
The positive direction of electrode potential, thus resembling noble

metals such as gold and platinum.
Noble Metal:
(1) A metal that occurs commonly in nature in the free state.

(2) A metal or alloy whose corrosion products are formed with a small
negative or a positive free-energy change.
Noble Potential:
A metal is said to exhibit a noble potential when it is cathodic to other

metals. Copper is noble compared to magnesium. This term is
analogous to electro-positive.
Ogive:
An X-Y plot line graph of a cumulative frequency or cumulative

relative frequency distribution often presented in percentages. An
ogive has the following components:
A title, which identifies the population or sample.
A vertical scale, which identifies either the cumulative frequencies or
the cumulative relative frequencies.
A horizontal scale, which identifies the upper class boundaries. The
horizontal scale for an ogive is always based on the upper class
boundaries.
The main use of an ogive is to estimate percentiles. The twentieth
percentile is the value below which 20 percent of the data falls.
Important percentiles are the median (50 percent), lower quartile (25
percent), and upper quartile (75 percent).
191
Open Circuit Potential:
The measured potential of a cell from which no current flows in the

external circuit. Can also be defined as the potential differential
between two separate poles.
Organic Zinc-Rich
Coating:
Coating containing a metallic zinc pigment (typically 75 wt% zinc or

more in the dry film) in an organic resin.
Oxidation:
(1) Loss of electrons by a constituent of a chemical reaction.

(2) Corrosion of a metal that is exposed to an oxidizing gas at elevated
temperatures.
Oxidation-Reduction
Potential:
The potential of a reversible oxidation-reduction electrode measured

with respect to a reference electrode, corrected to the hydrogen
electrode, in a given electrolyte.
Paint:
A pigmented liquid or resin applied to a substrate as a thin layer that is

converted to an opaque solid film after application. It is commonly
used as a decorative or protective coating.
Passivation:
The process or processes by which a metal becomes inert to a given

environment or environments.
Passive:
(1) The positive direction of electrode potential.

(2) A state of a metal in which a surface reaction product causes a
marked decrease in the corrosion rate relative to that in the absence of
the product.
pH:
A measure of the acidity or alkalinity of a solution. A value of seven is

neutral; low numbers are acid, large numbers are alkaline. Strictly
speaking, pH is the negative logarithm of the hydrogen ion
concentration.
Pitting:
Cavities formed in a material caused by localized corrosion.
Polarization:
The change of electrode potential resulting from the effects of current

flow, measured with respect to steady-state potentials.
Pozzolan:
A pozzolan is a finely divided material that reacts with calcium

hydroxide and alkalies to form compounds possessing cementitious
properties.
Primer:
A coating material intended to be applied as the first coat on an

uncoated surface. The coating is specifically formulated to adhere to
and protect the surface as well as to produce a suitable surface for
subsequent coats.
192
Protective Potential:
A term sometimes used in cathodic protection to define the minimum

potential required to suppress corrosion.
P/S:
A measurement of the difference in potential between a structure and a

copper-copper sulfate half-cell electrode in contact with an electrolyte,
soil or water.
Reduction:
Gain of electrons by a constituent of a chemical reaction, occurs at

cathode.
Resistivity:
The relative degree to which soil or water resists the flow of electric
current. The most common terms in use are the ohmmeter and the
ohm-centimeter.
Rust:
Corrosion product consisting primarily of iron oxide. A term properly

applied only to iron and other ferrous metals.
Sacrificial Protection:
Reduction or prevention of corrosion of a metal in an environment by

coupling it to another metal which is electrochemically more active in
that particular environment.
Short:
An inadvertent, undesirable contact between two buried metals, or the

electrical failure of installed insulation which destroys the desired
metallic isolation or a system.
Soil Potential Gradient: The voltage drop in the soil caused by direct current flowing away
from or to a ground electrode, anode or cathode. The voltage gradient
is measured between two copper-copper sulfate half-cell electrodes
D distance apart on a radius line from the ground electrode.
Stray Current
Corrosion:
Corrosion caused by direct current flowing through paths other than

the intended circuit; for example, corrosion caused by current
originating from direct current transportation or direct current
transmission systems.
Stress Corrosion
Cracking:
Cracking of a material produced by the combined action of corrosion

and tensile stress (residual or applied).
193
Throw:
The polarization that occurs as the result of a rectifier station is

generally greatest at the anode bed and decreases along the protected
structure. In the case of an impressed current rectifier station that
produces a direct current output, the protection "attenuates" along the
electrically continuous segments of the structures. In the case of pulse
rectifier stations that produce short pulses of high voltage direct
current, the protection will not theoretically be limited by electrical
discontinuities along the pipeline and the distance protection is
achieved is referred to as "throw."
Throwing Power:
The relationship between the current density at a point on a surface

and its distance from the counterelectrode. The greater the ratio of the
surface resistivity shown by the electrode reaction to the volume
resistivity of the eletrolyte, the better is the throwing power of the
process.
Void:
(1) A holiday, hole, or skip in a coating.

(2) A hole in a casting or weld deposit usually resulting from
shrinkage during cooling.
White Metal Blast

Cleaned Surface:
A white metal blast cleaned surface, when viewed without

magnification, shall be free of all visible oil, grease, dust, dirt, mill
scale, rust, coating, oxides, corrosion products, and other foreign
matter.
Wire Drawing:
The art or process of wire drawing is to pull or draw a wire of a bigger

diameter through a hole with smaller diameter thereby reducing the
diameter through plastic deformation while the volume remains the
same.
194
REFERENCES AND BIBLIOGRAPHY

Almughery, A., P. Gaydcki, M. Zaid, B. Fernandes, G. Miller, F. Elmandaani, and O. Wojvola
(2005), Inductive Scan Imaging Technique in Detecting and Visualizing Corrosion for
Large-Scale Prestressed Concrete Cylinder Pipes, Pipelines 2005, ASCE, Reston, VA, pp.
980-988.
American Concrete Pressure Pipe Association (1997), 1997 Pipeline Operations and
Maintenance Cost Study.
ANSI/AWWA C301-06 (Draft), Standard for Prestressed Concrete Pressure Pipe, Steel
Cylinder Type, American Water Works Association, Denver, CO, Foreword.
ASTM A82-34, Standard Specifications for Cold-Drawn Steel Wire for Concrete Reinforcement.
ASTM A227-47, Standard Specifications for Hard-Drawn Steel Spring Wire.
ASTM A229-41, Standard Specifications for Oil-Tempered Steel Spring Wire.
ASTM A648-72, Standard Specification for Steel Wire, Hard Drawn for Prestressing Concrete
Pipe, Philadelphia, PA.
ASTM C33, Standard Specification for Concrete Aggregates.
ASTM C127, Standard Test Method for Density, Relative Density (Specific Gravity), and
Absorption of Coarse Aggregate.
ASTM C128, Standard Test Method for Density, Relative Density (Specific Gravity), Absorption
of Fine Aggregate.
ASTM C494, Standard Specification for Chemical Admixtures for Concrete.
ASTM D75, Standard Practice for Sampling Aggregates.
AWWA C301-07, Proposed Revisions to AWWA Standard for Prestressed Concrete Pressure
Pipe for Water and Other Liquids, American Water Works Association Standards
Committee on Concrete Pressure Pipe, Denver, CO.
AWWA C301-99, AWWA Standard for Prestressed Concrete Pressure Pipe, Steel-Cylinder
Type, American Water Works Association, Denver CO.
AWWA C301-92, AWWA Standard for Prestressed Concrete Pressure Pipe, Steel-Cylinder
Type, for Water and Other Liquids, American Water Works Association, Denver CO.
AWWA C301-84, AWWA Standard for Prestressed Concrete Pressure Pipe, Steel Cylinder
Type, for Water and Other Liquids, American Water Works Association, Denver, CO.
Type, for Water and Other Liquids, American Water Works Association, Denver, CO.
Type, for Water and Other Liquids, American Water Works Association, New York, NY.
AWWA C301-64, AWWA Standard for Reinforced Concrete Water Pipe Steel Cylinder Type,
Prestressed, American Water Works Association, New York, NY.
AWWA C301-58, AWWA Standard for Reinforced Concrete Water Pipe Steel Cylinder Type,
Prestressed, American Water Works Association, New York, NY.
AWWA C301-55T, Tentative AWWA Standard for Reinforced Concrete Water Pipe-Steel
Cylinder Type, Prestressed, American Water Works Association, New York, NY.
AWWA C301-52, Standard Specifications for Reinforced Concrete Water Pipe-Steel Cylinder
Type, Prestressed, American Water Works Association, New York, NY.
AWWA 7B.2-T (1949), Tentative Standard Specifications for Reinforced Concrete Water Pipe
Steel Cylinder Type, Prestressed, distributed by Price Brothers Company, Dayton, OH.
195
AwwaRF (2005), External Corrosion and Corrosion Control of Buried Water Mains (90987,
2005), Denver, CO.
AwwaRF (2005), Risk Management of Large Water Transmission Mains (91087, 2005), Denver,
CO.
AwwaRF (1995), Performance of Prestressed Concrete Pipe (unpublished.) pg 200.
Baik Shik, Hyeon; Dulay M. Abraham; and Dean J. Gipson (2004), Impacts of Deterioration
Based Life Cycle Cost Analysis (LCCA) on the Assessment of Values of Wastewater
Infrastructure Assets, Pipeline Engineering and Construction Whats on the Horizon?
John J. Galleher and Michael Stift, eds., ASCE, Reston, VA (CD-ROM).
Bambei, John H., Jr., and Richard Lewis (2005), Correlation of Wire Breaks on Prestressed
Concrete Cylinder Pipe with Predictions from Electromagnetic Testing, Pipelines 2005,
ASCE, Reston, VA, pp. 989-1000.
Bell, Graham E. C.; Donald R. Kendall; and Susan Mulligan (2001), An Assessment and
Reliability Program for PCCP, Pipelines 2001-Advances in Pipeline Engineering &
Construction, Joseph Castronovo, ed., ASCE, Reston, VA (CD-ROM).
Bengtsson, Brad; J. Koppenhaver; B. Mergalas; X. Kong; and A. Lee (2005), Next Steps in the
Evolution of an Acoustic Emission Testing System, Pipelines 2005, ASCE, Reston, VA, pp.
1025-1034.
Bichler, Robert (2005), Evaluation and Management of Prestressed Concrete Cylinder Pipe
Assets, Pipelines 2005, ASCE, Reston, VA, pp. 116-129.
Bradish, Bryan M.; R. J. Cronin; and R. O. Lewis (1995), When Should PCCP with Interpace
Class IV Wire Be Replaced?, Advances in Underground Pipeline Engineering, J. Jeyapalan,
ed., ASCE, NY, NY, pp. 156-167.
Cromwell, John E., III; Haydn Reynolds; Norman Pearson, Jr.; and Michael Grant (2002), Costs
of Infrastructure Failure, Awwa Research Foundation, Denver, CO.
Deb, Arun K.; Jerry K. Snyder; James J. Chelius; Jack Urie; and D. Kelly ODay (1990),
Assessment of Existing and Developing Water Main Rehabilitation Practices, Awwa
Research Foundation, Denver, CO.
Diab, Youssef G. and Thomas Bonierbale (2001), A Numerical Modeling and a Proposal for
Rehabilitation of PCCPs, Pipelines 2001-Advances in Pipeline Engineering &
Diaz, Robert; D. Campbell; and M. Holley (2005), Acoustic Monitoring and Replacement of a
Distressed 42-Inch Prestressed Concrete Transmission Main, Pipelines 2005, ASCE,
Reston, VA, pp. 390-400.
Doniguian, Ted; H. Kips; and J. Barnes (1998), Extending the Life of Prestressed Concrete
Cylinder Pipe With Pulse Cathodic Protection, Pipelines in the Constructed Environment,
Duranceau, Steven J., Melvin J. Schiff, and Graham E. C. Bell (1996), Effects of Electrical
Grounding on Pipe Integrity and Shock Hazard, AwwaRF, Denver, CO, ISBN 0-89867861.
Essamin, Omar; K. El-Sahli; G. Hovanessian; and T. LeDiouron (2005), Risk Management
System for Prestressed Concrete Cylinder Pipeline: Practical Results and Experience on
the Great Man Made River, Pipelines 2005, ASCE, Reston, VA, pp. 241-251.
196
Fiori, Steven R.; Donald R. Kendall; and Susan B. Mulligan (2001), Rehabilitation of
Prestressed Concrete Cylinder Pipe A Utility Perspective, Pipelines 2001-Advances in
Pipeline Engineering & Construction, Joseph Castronovo, ed., ASCE, Reston, VA (CDROM).
Galleher, John J.; Graham E. C. Bell; and Andrew Romer (2005), Comparison of Two
Electromagnetic Techniques to Determine the Physical Condition of PCCP, Pipelines
2005, ASCE, Reston, VA, pp. 401- 410.
Galleher, John J., Jr.; Gary P. Stine; Michael K. Kenny; and Michael Stift (2001), Advances in
Construction Documentation for Future Pipeline Condition Assessments, Pipelines 2001Advances in Pipeline Engineering & Construction, Joseph Castronovo, ed., ASCE, Reston,
VA (CD-ROM).
Galleher, John J. and Michael T. Stift (1998), Internal Inspection and Database Development of
PCCP, Pipelines in the Constructed Environment, ASCE, Reston, VA, pp. 721-730.
Gomez, Roberto; David Muoz; Ricardo Vera; and Jose A. Escobar (2004), Structural Model for
Stress Evaluation of Prestressed Concrete Pipes of the Cutzamala System, Pipeline
Engineering and Construction Whats on the Horizon? John J. Galleher and Michael
Stift, eds., ASCE, Reston, VA (CD-ROM).
Heger, F. J.; M. S. Zarghamee; and W. R. Dana (1990), Limit States Design of Prestressed
Concrete Cylinder Pipe, Part I: Criteria, Journal of Structural Engineering, 116:8:20832104.
Henry, Gregory J.; J. Miller; and S. Do (2005), Fast-Track Repair of 42-Inch PCCP Water
Main, Houston, Texas, Pipelines 2005, ASCE, Reston, VA, pp. 161-171.
Jackson, Rodney Z.; C. Pitt; and R. Skabo (1992), Nondestructive Testing of Water Mains for
Physical Integrity, Awwa Research Foundation, Denver, CO.
Khondiker, Sufian A. and John R. Mitchell, Jr. (1998), Design and Construction Aspects of the
PCCP Water Main Rehabilitation, Pipelines in the Constructed Environment, ASCE,
Kleiner, Yehuda; Balvant Rajani; and Rohan Sadiq (2005), Risk Management of Large-Diameter
Water Transmission Mains, Awwa Research Foundation, Denver, CO.
Kleiner, Yehuda; Rehan Sadiq; and Balvan Rajani (2004), Modeling Failure Risk in Buried
Pipes Using Fuzzy Markov Deterioration Process, Pipeline Engineering and Construction
Whats on the Horizon? John J. Galleher and Michael Stift, eds., ASCE, Reston, VA
(CD-ROM).
Knowles, William L. C. (1990), Failure of Prestressed Concrete Embedded Cylinder Pipe,
Design and Installation, Kenneth Kienow, ed., ASCE, NY, NY, pp. 434-441.
Kong, Xiangjie and Brian Mergalas (2005), Condition Assessment of Small Diameter Water
Transmission Mains, Pipelines 2005, ASCE, Reston, VA, pp. 252-262.
Lewis, Richard A. and Paul Fisk (2005), Detection of Prestressed Concrete Cylinder Pipe
Thinning From Hydrogen Sulfide Deterioration.
Lillie, Kevin; Christopher Reed; Mark A. R. Rodgers; Simon Daniels; and David Smart (2004),
Workshop on Condition Assessment Inspection Devices for Water Transmission Mains,
Awwa Research Foundation, Denver, CO.
Lofti, Hamid R.; R. Oesterle; and J. Roller (2005), Reliability Assessment of Distressed
Prestressed Concrete Cylinder Pipe, Pipelines 2005, ASCE, Reston, VA, pp. 838-852.
197
Lynch, Brendan; Thomas Y. Wilson; and James P. Gunther (2004), Ten Times in Ten Years,
Pipeline Engineering and Construction Whats on the Horizon? John J. Galleher and
Michael Stift, eds., ASCE, Renton, VA (CD-ROM).
Marshall, David H.; M. Zarghamee; B. Mergalas; and Y. Kleiner (2005), Tarrant Regional
Water Districts Risk Management Plan for Prestressed Concrete Cylinder Pipe, Pipelines
2005, ASCE, Reston, VA, pp. 853-861.
Marshall, David; Mike Meza; and Sandy Swinnea (2002), Managing Maintenance Costs of a
Water Transmission System, Pipelines 2002-Beneath Our Feet: Challenges and Solutions,
George E. Kurz, ed., ASCE, Reston, VA (CD-ROM).
Marshall, David H. (1998), TRWD Experience with Prestressed Concrete Pipe, Pipelines in the
Constructed Environment, ASCE, Reston, VA, pp. 556-565.
McCaffrey, K.; D. R. Kendall; and S. Mulligan (2006), Steel Liners in Large Diameter
Prestressed Concrete Cylinder Pipe: Calleguas Municipal Water Districts 2005 Failure,
Repair, Evaluation, and Reinforcement, Proceedings, ASCE Pipelines 2006 Conference,
Service to the Owner.
Mergalas, Brian; X. Kong; D. Roy; and J. Balliew (2005), Using a Combination of Condition
Based Asset Management Techniques to Manage a Water Transmission System, Pipelines
2005, ASCE, Reston, VA, pp. 782-792.
Mergalas, Brian and Xiangjie Kong (2001a), Electromagnetic Inspection of Prestressed
Concrete Pressure Pipe, Awwa Research Foundation, Denver, CO.
Mergalas, Brian J.; D. L. Atherton; and X. Kong (2001b), NDE Inspection of PCCP Using
Remote Field Eddy Current/Transformer Coupling, Pipelines 2001-Advances in Pipeline
Engineering & Construction, Joseph Castronovo, ed., ASCE, Reston, VA (CD-ROM).
Mergalas, Brian J.; D. L. Atherton; R. Konyalian; and M. Zarghamee (2001c), Managing the
Risk in Operating PCCP Pipelines, Pipelines 2001-Advances in Pipeline Engineering &
Mergalas, Brian J. and David L. Atherton (1998), In-Line Electromagnetic Inspection of PCCP,
Pipelines in the Constructed Environment, ASCE, Reston, VA, pp. 714-720.
Moncreif, W. Jeffrey; Donald R. Kendall; Susan B. Mulligan; Robert C. Blake; and Craig M.
Watkins (2001), Rehabilitation of 78-Inch PCCP with Carbon Fiber Reinforced Composite
Material, Pipelines 2001-Advances in Pipeline Engineering & Construction, Joseph
Castronovo, ed., ASCE, Reston, VA (CD-ROM).
Najafi, Mohammad and Guru Kulandaivel (2005), Pipeline Condition Prediction Using Neural
Network Models, Pipelines 2005, ASCE, Reston, VA, pp. 767-781.
Nelson, Richard E. (2005), Utilizing Condition Assessment Data in Pipeline Replacement and
Reinvestment Planning, Pipelines 2005, ASCE, Reston, VA, pp. 130-136.
ODay, D. Kelly; R. Weis; S. Chiaviari; and D. Blair (1986), Water Main Evaluation for
Rehabilitation/Replacement, Awwa Research Foundation, Denver, CO.
Ojdrovic, Rasko P.; M. S. Zarghamee; J. R. Hegar; and T. Westman (2001), Condition
Assessment of a PCCP Line Accessible from Outside Only, Pipelines 2001-Advances in
Pipeline Engineering & Construction, Joseph Castronovo, ed., ASCE, Reston, VA (CDROM).
Olander, H. C. (1950), Stress Analysis of Concrete Pipe, Engineering Monograph No. 6, U.S.
Bureau of Reclamation, Department of the Interior, Washington, DC.
Ortega, Rafael; G. Henry; and H. Hovsepian (2005), Pipeline Ruptures: High Demand for Water
Prompts an Innovative Solution, Pipelines 2005, ASCE, Reston, VA, pp. 439-450.
198
Paris, J. H. (1921), Coefficients for Large Horizontal Pipes, Engineering News-Record, Vol.
87, p. 768.
Parks, Randall R.; J. K. Drager; and R. P. Ojdrovic (2001), Condition Assessment and
Rehabilitation of the Windy Gap Pipeline An Owners Perspective, Pipelines 2001Advances in Pipeline Engineering & Construction, Joseph Castronovo, ed., ASCE, Reston,
VA (CD-ROM).
Price, Robert E.; Richard A. Lewis; and Bernard Erlin (1998), Effects of Environment on the
Durability of Prestressed Concrete Cylinder Pipe, Pipelines in the Constructed
Environment, ASCE, Reston, VA, pp. 584-593.
Price, Robert E. (1990), The Investigation, Cause and Prevention of PCCP Failures,
proceedings, AWWA Annual Conference, pp. 663-681.
Prosser, David P. (1996), Research, Product Improvement, New AWWA Standards, Compliance
Certification & Performance Contribute to the Increased Usage of Prestressed Concrete
Cylinder Pipe, presented at the 1996 American Water Works Association Annual
Convention and Exhibition, June 25, 1996, Toronto, Ontario, Canada.
Reed, Christopher; Alastair J. Robinson; and David Smart (2004), Techniques for Monitoring
Structural Behaviour of Pipeline Systems, Awwa Research Foundation, Denver, CO.
Romer, Andrew E. and Graham E. C. Bell (2004), AwwaRF, External Corrosion and Corrosion
Control of Buried Water Mains, Denver, CO.
Stine, Gary P. and Michael T. Stift (1998), Rehabilitation of a 183 cm PCCP with Steel Plate
Liners, Pipelines in the Constructed Environment, ASCE, Reston, VA, pp. 594-601.
Suydam, Tim; James Woods; Ed Stewart; and Michael T. Stift (2001), Custom Built Installation
Equipment for Relining Prestressed Concrete Cylinder Pipe Minimizes Construction
Schedule, Pipelines 2001-Advances in Pipeline Engineering & Construction, Joseph
Castronovo, ed., ASCE, Reston, VA (CD-ROM).
von Fay, Kurt F. and Michael T. Peabody (1994), Historical Performance of Buried Water Pipe
Lines, U.S. Department of the Interior, Bureau of Reclamation, Denver, CO.
Villalobos, Jose L. (1998), Effects of Prestressed Concrete Cylinder Pipe in a High Chloride
Environment after 19 Years of Service, Pipelines in the Constructed Environment, ASCE,
Walsh, Terry L. and David S. Hodge (1998), Overcoming the Challenges of Replacing 20 km of
Defective 1524 mm Diameter PCCP, Pipelines in the Constructed Environment, ASCE,
Worthington, Will and William J. DiMarco (1996), Acoustic Monitoring to Enhance Pipeline
Safety at Crossings, Pipeline Crossings 1996, Lawrence F. Catalano, ed. ASCE, NY, NY,
pp. 1-13.
Zarghamee, Mehdi S., and Rasko P. Ojdrovic (2001), Risk Assessment and Repair Priority of
PCCP with Broken Wires, Pipelines 2001 - Advances in Pipeline Engineering and
Construction, Joseph P. Castronovo, ed., ASCE , Reston, VA, 2001.
Zarghamee, Mehdi S.; Rasko P. Ojdrovic; and Roger Fongemie (1998), Condition Assessment
and Repair of Prestressed Concrete Pipe, Pipelines in the Constructed Environment,
Zarghamee, Mehdi S. and W. R. Dana (1991), A Step-by-Step Integration Procedure for
Computing State of Stress in Prestressed Concrete Pipe, Computer Analysis of the Effects
of Creep, Shrinkage, and Temperature Changes on Concrete Structures, Pub. SP-129, ACI,
pp. 155-170.
199
Zarghamee, Mehdi S. (1990a), Evaluation of Combined Load Tests of Prestressed Concrete

Cylinder Pipe, Pipeline Design and Installation, Kenneth K. Kienow, ed., ASCE, NY,
NY, pp. 582-585.
Zarghamee, Mehdi S. and K. L. Fok (1990b), Analysis of Prestressed Concrete Pipe Under
Combined Loads, Journal of Structural Engineering, 116:7:2022-2039.
Zarghamee, Mehdi S.; K. L. Fok; and E. S. Sikiotis (1990c), Limit States Design of Prestressed
Concrete Cylinder Pipe, Part II: Procedure, Journal of Structural Engineering,
116:8:2105-2126.
200
ABBREVIATIONS
a-c
Alternating current
ASCE
American Society of Civil Engineers
ASME
American Society of Mechanical Engineers
ASTM
American Society for Testing and Materials
AWWA
American Water Works Association
AwwaRF
CCMP
Corrosion control master plan
CCP
Concrete cylinder pipe
CSE
Copper-copper sulfate reference electrode
d-c
Direct current
DWR
Department of Water Resources
EC-PCCP
Embedded-cylinder type prestressed concrete cylinder pipe
ECS
Electromagnetic conductivity surveys
EIS
Electrochemical impedance spectroscopy
EMF
Electromotive force
ENR
Engineering-News Record
GASB
Governmental Accounting Standards Board
GDP
Gross domestic product
GIS
Geographic information systems
GNP
Gross National Product
GPS
Global Positioning System
HE
Hydrogen embrittlement
ICCP
Impressed current cathodic protection
IEEE
Institute of Electrical Engineers
IRC
Institute for Research in Construction
201
JLF
Joint leak frequency
LC-PCCP
Lined-cylinder type prestressed concrete cylinder pipe
LF
Lineal feet
LPR
Linear polarization resistance
LWC
Louisville Water Company
ma
Milliampere
MBF
Main break frequency
ml
Milliliters
MRRP
Main replacement and rehabilitation program
mV
Millivolts
MWDSC
Metropolitan Water District of Southern California
NACE
National Association of Corrosion Engineers
NBS
National Bureau of Standards
NDT
Nondestructive testing
NEC
National Electrical Code
NIST
National Institute of Science and Technology
OD
Outside diameter
PAC
Project Advisory Committee
PCCP
Prestressed concrete cylinder pipe
PPIC
Pressure Pipe Inspection Company
ppm
Parts per million
PRCP
Pulsed rectifier cathodic protection
P/S
Pipe to soil
psi
Pounds per square inch
QA/QC
Quality assurance/quality control
202
RCPP
Reinforced concrete pressure pipe
RFEC
Remote field-eddy current
RF/TC
Remote field/transformer coupling
SCADA
Supervisory control and data acquisition
SCVWD
Santa Clara Valley Water District
SDCWA
San Diego County Water Authority
SSPC
Steel Structures Painting Council
TC
Transformer coupling
UL
Underwriters Laboratory
USBR
U.S. Bureau of Reclamation
203
Figure 2.1 PCCP Timeline

Source: Failure of Prestressed Concrete Cylinder Pipe by Romer et al. 2008 AwwaRF. ALL RIGHTS RESERVED.
6666 West Quincy Avenue

Denver, CO 80235-3098 USA
P 303.347.6100
www.awwarf.org
email: info@awwarf.org
Sponsors Research
Develops Knowledge
Promotes Collaboration
1P-3.5C-91214-05/08-NH

Failure of Prestressed Concrete Cylinder Pipe

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Failure of Prestressed Concrete Cylinder Pipe

Hochgeladen von

Copyright:

Verfügbare Formate

Failure of Prestressed

Concrete Cylinder Pipe

2008 AwwaRF. ALL RIGHTS RESERVED

About the Awwa Research Foundation

2008 AwwaRF. ALL RIGHTS RESERVED

Jointly sponsored by:

2008 AwwaRF. ALL RIGHTS RESERVED

2008 AwwaRF. ALL RIGHTS RESERVED

Reported failures of prestressed concrete cylinder pipe ......................................................5

Differences between lined and embedded type PCCP.......................................................53

PCCP significant changes with time..................................................................................53

Production of prestressed concrete cylinder pipe ..............................................................56

AWWA C301 prestressed concrete cylinder pipe timeline ...............................................61

Minimum wire clearance for minimum wire size..............................................................66

Joint dimensions ................................................................................................................67

Timeline of wire standards for PCCP ................................................................................69

Summary statistics ...........................................................................................................129

Weibull summary statistics ..............................................................................................136

2008 AwwaRF. ALL RIGHTS RESERVED

Location of failures LC-PCCP........................................................................................51

Location of failures EC-PCCP........................................................................................52

Timeline ................................................................................................................. see insert

Total production PCCP by manufacturer.....................................................................76

Total production PCCP- ECP compared to LCP ............................................................77

Total ECP production by manufacturer ..........................................................................78

Total LCP production by manufacturer ..........................................................................79

Minimum cylinder thickness PCCP................................................................................80

Minimum wire size PCCP ..............................................................................................80

Mortar coating PCCP......................................................................................................81

Lifespan histogram all samples.......................................................................................95

Lifespan histogram installed pre-1955 ...........................................................................96

Lifespan histogram installed 1955-1963.........................................................................97

Lifespan histogram installed 1964-1967.........................................................................98

Lifespan histogram installed 1968-1971.........................................................................99

Lifespan histogram installed 1972-1978.......................................................................100

Lifespan histogram installed 1972-1978: Interpace......................................................101

Lifespan histogram installed 1972-1978: unknown and non-Interpace........................102

Lifespan histogram installed 1979-1991.......................................................................103

Lifespan histogram installed 1992-2007.......................................................................104

Failures by installation date histogram Category 1, 1942-2006 ...................................105

Failures by installation date with production histogram Category 1, 1942-2006.........106

Failures by installation date histogram Categories 2 and 3, 1942-2006.......................107

Normalized failures by installation date histogram Category 1, 1942-2006 ................109

Normalized failures by installation date histogram Categories 2 and 3,

Failure rates by time period .............................................................................................111

Failure rates by wire class histogram...............................................................................112

Failure rates by pipe type by year installed histogram Categories 2 and 3,

Lifespan ogive all samples...............................................................................................138

Lifespan ogive installed pre-1955....................................................................................139

Lifespan ogive installed 1955-1963.................................................................................140

Lifespan ogive installed 1964-1967.................................................................................141

Lifespan ogive installed 1968-1971.................................................................................142

Lifespan ogive installed 1972-1978.................................................................................143

Lifespan ogive installed 1972-1978: Interpace................................................................144

Lifespan ogive installed 1972-1978: unknown and non-Interpace..................................145

Lifespan ogive installed 1979-1991.................................................................................146

Lifespan ogive installed 1992-2007.................................................................................147

Probability Weibull: all samples (Category 1) .............................................................148

Probability Weibull: all samples (Categories 1 and 2) .................................................149

Probability Weibull: all samples (Category 1 and Categories 2 and 3)........................150

Probability Weibull: all samples (Category 3) .............................................................151

Probability Weibull: installed pre-1955 (Category 1) ..................................................152

Probability Weibull: installed pre-1955 (Cateogries 1 and 2) ......................................153